Lecture 1.

pdf
Lecture 2.pdf
Lecture 3.pdf
Lecture 4.pdf
Lecture 5.pdf
Lecture 6.pdf
Lecture 7.pdf
Lecture 8.pdf
Lecture 9.pdf
Lecture 10.pdf
Lecture 11.pdf
Lecture 12.pdf
Lecture 13.pdf
Lecture 14 .pdf
 SAS Components

This class
Base SAS – basic procedures and data management

SAS has over 200 components, including

SAS/STAT – statistical analysis
SAS/GRAPH – high quality graphics & presentations
SAS/ACCESS – reads data directly from databases
SAS/ETS – econometrics and time series
SAS/INSIGHT – data mining
SAS/QC – quality control
SAS/PH – clinical trials
 Applications

Why are we using it?

What are some of the applications?

Statistical analysis

Data management

Creating reports
 Today’s Objectives

Open the program

Learn how to navigate the program

Learn some basic syntax

Create our first SAS program!

 First Step

Open SAS
Menu -> All Programs -> Math & Stats -> SAS -> SAS 9.4
 The Main Windows

Five Main Windows

Editor

Log

Output

Results

Explorer
 Editor

Edit, execute, and save SAS programs

 Log

Displays status messages regarding the execution of

SAS procedures
 Results

Displays links to previously executed results

 Explorer

Navigate libraries, data sets, and other SAS objects

 Output + Results Viewer

Output window:

Results viewer
Default output for SAS 9.4
A single, continuous html report
Not affected by options like
Page size, page number, etc.
Another Useful Window: Table Editor
 Getting Started

How to submit/execute a program?

Make sure the Editor window is active
Submitting the entire program
Run Submit
Click on the “running man” symbol in the toolbar

Command line Type ‘submit’ and hit Enter

F8
Submitting a portion of the program
Highlight the portion you’d like to submit
• <right click> Submit Selection
• F8
 Getting Started

How to save a program?

File Save
Ctrl + S
 File Extensions

What are the file extensions for

A SAS program?
.sas
A SAS data set?
.sas7bdat
A SAS log file?
.log
A SAS output file?
.html (by default)
 The Basics: SAS Process

Input Refine Statistical Generate

Data Data Analysis Output
 The Basics

SAS statements
Always begin with a keyword
Always end with a semicolon (;)
Are free format
i.e. Can begin at any location and end at any location
Entire program can be written on one line, or many lines
EXCEPT when using the datalines; statement
SAS
Is not case sensitive
i.e. daTa nOtCaseSensitive;
EXCEPT in the case of string comparisons
 The Basics

Names of SAS data sets and variables must

Be no longer than 32 characters

Begin with a letter or underscore

Contain only letters, numbers, or underscores (_)

 The Basics

Comments

What are they?

Why should we use them?

Single line: begin with an asterick (*) and ends with a

semicolon (;)

Multiple line: begins with a /* and ends with a */

 The Basics

SAS 9.4 has context-sensitive help.

Highlight a keyword and press F1
 Libraries

SAS file names are always contain 2 levels

Level 1: <library-name>
Level 2: <data-set-name>
i.e. <library-name>.<data-set-name>

What does this mean?

SAS references folders called libraries when accessing SAS
data sets
Libraries are simply pointers to folder locations on the disk
drive
i.e. ‘X:\PStat 130\data1’
 Libraries

SAS has 2 existing libraries:

work
sasuser
work
Is a temporary library
Is the default library
sasuser
Is a permanent library

All other libraries must be assigned

 Libraries

Library references (libref) must

Start with a letter or an underscore
Be 8 characters or less

To define a library, you need

The libname keyword
A user-defined libref name
A folder location

General format:
libname desktop 'C:\desktop';
 The DATA Step

DATA Step: A set of statements that

read in a data file
assign variable names, labels, and formats
select specific observations
Data set being External file
created containing raw data
Keyword DATA NewData;
“DATA”
infile 'C:\rawdata.txt';
input name $ age;
run;

Variables being created

End of DATA Step in SAS data set
 The PROC Step

PROC Step: A set of statements that

perform “utility” operations on a data set
analyze data
output results or reports
Procedure being SAS data set to
executed be summarized
Keyword “PROC” PROC freq data=NewData;
tables age;
run;

Variable being
End of PROC Step Keyword “Tables” summarized
 First SAS Program

DATA intelligence;
input IQ;
datalines;
99 i.e. work.intelligence
140
125
118
104
;
run;

PROC print;
run;
William Qiu
Department of Statistics and Applied Probabilty
UCSB
PSTAT 130 - Summer 2017 William Qiu 2
Objectives
 Create a printed report using PROC PRINT
 Learn PROC PRINT syntax and options
 Define titles and footnotes to enhance reports
 Define descriptive column headings
 Use SAS system options

PSTAT 130 - Summer 2017 William Qiu 3

PSTAT 130 - Summer 2017 William Qiu 4
PROC PRINT
The PRINT procedure prints the observations in a SAS data set, using all or
some of the variables. You can create a variety of reports ranging from a simple
listing to a highly customized report that groups the data and calculates totals
and subtotals for numeric variables.
 Simple Version
PROC PRINT will only let you print a single data set at a time

proc print;
run;
 With Data File
proc print data=ia.empdata;
run;

With Variable Selection: identifies the variables to print. PROC PRINT prints
the variables in the order that you list them.
proc print data=ia.empdata;
var JobCode EmpID Salary;
run;
PSTAT 130 - Summer 2017 William Qiu 5
Proc Print Example
Libname ia <insert folder reference>

proc print data=ia.empdata;

var JobCode EmpID Salary;
run;

PSTAT 130 - Summer 2017 William Qiu 6

PROC PRINT Default Output
 Use variable names as column headings
 Question: how to change variable names?
 Eg. Month->Mon
 Display all variables contained in data set
 Question: how to select variables that we are interested in?
 Display all observations contained in data set
 Question: how to select a subset of observations that we want
to display?
 Display variable values in their “native” format
 Question: how to change formats of variables? Eg. Decimal
places

PSTAT 130 - Summer 2017 William Qiu 7

Defining Titles and Footnotes
Control the content, apperance and placement of title and footnote text.
 General form of the TITLE statement
TITLEn ‘text’
Syntax: TITLE <n> <ods-format-options> <'text' | "text">;
n: specifies the relative line that contains the title line.
ods-format-options specifies formatting options for the ODS HTML, RTF, and PRINTER
destinations., eg. Color=red, height=, (custinuzer and enhance your ooutput)
'text' | "text" specifies text that is enclosed in single or double quotation marks.

 General form of the FOOTNOTE statement

FOOTNOTEn ‘text’;

 Examples
 title1 ‘PSTAT 130 Homework #1’; suppresses a title on line 1 and all lines
after it:
 footnote2 ‘Confidential’;

PSTAT 130 - Summer 2017 William Qiu 8

Defining Titles and Footnote
Features of titles:
 Titles appear at the top of the page of output
 The default title is “The SAS System”
 You can have more than title line
 The value of n can be from 1 to 10 and refers to title line
 An unnumbered TITLE is equivalent to TITLE1
 Titles remain in effect until they are changed
 The null TITLE statement , title; cancels all titles

PSTAT 130 - Summer 2017 William Qiu 9

Defining Titles and Footnote
Features of footnotes:
 Footnotes appear at the bottom of the page of output
 There is no default footnote (different from title!)
 You can have more than footnote
 The value of n can be from 1 to 10 and refers to footnote
line
 An unnumbered FOOTNOTE is equivalent to FOOTNOTE1
 Footnotes remain in effect until they are changed
 The null FOOTNOTE statement, footnote; cancels all
footnote

PSTAT 130 - Summer 2017 William Qiu 10

Changing Titles and Footnotes
TITLEn or FOOTNOTEn
 replaces a previous title or footnote with the same
number
 Cancels all titles or footnotes with higher numbers

PSTAT 130 - Summer 2017 William Qiu 11

Defining Titles and Footnotes
PROC PRINT Code Resultant Title(s)
proc print data=work.march;
title1 ‘The First Line’;
title2 ‘The Second Line’;
run;
proc print data=work.april;
title2 ‘The Next Line’;
run;
proc print data=work.may;
title ‘The Top Line’;
run;
proc print data=work.june;
title3 ‘The Third Line’;
run;
proc print data=work.july; Cancel all previous titles or
title; footnotes, that are in
run; effect.

PSTAT 130 - Summer 2017 William Qiu 12

PSTAT 130 - Summer 2017 William Qiu 13
Assigning Column Labels
 Enhance your PROC PRINT report by labeling columns with
more descriptive text.
 General form of the LABEL statement
LABEL variable=‘label’
variable=‘label’;
Note: 1.labels can be up to 256 characters long. Enclose
the label in quotation marks.
2. The LABEL statement applies only to the PROC step in
which it appears.
 Examples:
LABEL Units=‘Number of Units’
DOB=‘Date of Birth’;

PSTAT 130 - Summer 2017 William Qiu 14

Assigning Column Labels
libname ia 'X:\PStat 130\data1'; (code may be varied)

ia.empdata

Default Variable Names

PSTAT 130 - Summer 2017 William Qiu 15

Assigning Column Labels
ia.empdata
This is also a
case of using
Multiple LABEL
statement to
assign multiple
names. See
next page for an
example.

PSTAT 130 - Summer 2017 William Qiu 16

PSTAT 130 - Summer 2017 William Qiu 17
Assigning Column Labels
ia.empdata Use the option
Split instead of
Label. Almost the
same.
Split option control
where text strings
split in labels.

PSTAT 130 - Summer 2017 William Qiu 18

Using SAS Options
 SAS system options are initialized with default settings when SAS is
invoked. However, the default settings for some SAS system options vary
both by operating environment and by site. Your on-site SAS support
personnel might have created a default options table. The default options
table is created by your site administrator and provides a global set of
default values that are specific for your site. It reduces the need to
duplicate options in every system configuration file at your site.

 General form of the OPTIONS statement:

OPTIONS option...;

 Run the following program to generate a list of all options (the list will
appear in the LOG window):
PROC OPTIONS;
RUN;

PSTAT 130 - Summer 2017 William Qiu 19

Using SAS System Options
Selected SAS system options:
 DATE vs. NODATE
 Determines whether SAS displays the system date on output
 LINESIZE=width (LS=width)
 Determines size of output line before line break
 PAGESIZE=n (PS=n)
 Determines number of lines on page before page break
 NUMBER vs. NONUMBER
 Determines whether SAS displays page numbers
 PAGENO=n
 Determines starting page number
 Example: (we are going to see more in homework 1)
Options nodate nonumber ls=72;

PSTAT 130 - Summer 2017 William Qiu 20

Formatting Data Values
In your SAS reports, formats control how the data values are displayed.
To make data values more understandable when they are displayed in
your procedure output, you can use the FORMAT statement, which
associates formats with variables.

Formats affects only how the data values appear in output, not the
actual data values as they are stored in the SAS data set.

PSTAT 130 - Summer 2017 William Qiu 21

Objectives
 Display formatted values using SAS formats in a list report
 Create user-defined formats using the FORMAT procedure
 Apply user-defined formats to variables in a list report

PSTAT 130 - Summer 2017 William Qiu 22

Using SAS Formats
an example for preview
Enhance the readability of reports by formatting the data
values

PSTAT 130 - Summer 2017 William Qiu 23

Using SAS Formats
Create custom formats to recode data values in a report

For a list of formats, see Formats, By Category in SAS Help

PSTAT 130 - Summer 2017 William Qiu 24

Formatting Data Values

SAS
Format Report
Data
Set

PSTAT 130 - Summer 2017 William Qiu 25

Formatting Data Values
 General form of the FORMAT statement
FORMAT variable(s) format;
Variable(s) is the name of one or more variables whose values are
to be written according to a particular pattern.
Format-name specifies a SAS format or a user-defined format that
is used to write out the values.
Note: the FORMAT statement applies only to the PROC step in which
it appears.
 Example:
proc print data=ia.empdata split=' ';
format Salary dollar11.2;
run;

PSTAT 130 - Summer 2017 William Qiu 26

What is a SAS Format?
<$> format<w>.<d> Number of
Indicates a decimal places
character
format
Format name Required
delimiter

Total width (including decimal

places and special characters)

PSTAT 130 - Summer 2017 William Qiu 27

SAS Formats
 Selected SAS formats

Format/ Result Description

Example
w.d standard numeric format
8.2 12234.21 Width=8, 2 decimal places
$w. standard character format
$5. KATHY width=5
COMMAw.d Commas in a number
COMMA9.2 12,234.21 Width=9, 2 decimal places
DOLLARw.d Dollar signs and commas in a number
DOLLAR10.2 $12,234.21 Width=10, 2 decimal places

PSTAT 130 - Summer 2017 William Qiu 28

SAS Formats
Stored Value Format Displayed
Value
27134.2864 COMMA12.2 27,134.29
27134.2864 12.2 27134.29
27134.2864 DOLLAR12.2 $27,134.29
27134.2864 DOLLAR9.2 $27134.29
27134.2864 DOLLAR8.2 27134.29
27134.2864 DOLLAR5.2 27134
27134.2864 DOLLAR4.2 27E3

PSTAT 130 - Summer 2017 William Qiu 29

Formatting Data Values
ia.empdata
Note the
difference
between the
original
format and
the new
format in the
output!

PSTAT 130 - Summer 2017 William Qiu 30

Date Formats
MMDDYYw. DATEw.
Format Displayed Format Displayed
Value Value
MMDDYY6. 101601 DATE7. 16OCT01
MMDDYY8. 10/16/01 DATE9. 16OCT2001
MMDDYY10. 10/16/2001

PSTAT 130 - Summer 2017 William Qiu 31

DATE Formats and Serial Date Values
Stored Value Format Displayed Value
0 MMDDYY8. 01/01/60
0 MMDDYY10. 01/01/1960
1 DATE9. 01JAN1960
-1 WORDDATE. December 31, 1959
365 DDMMYY10. 31/12/1960
366 WEEKDATE. Sunday, January 1,
1961

PSTAT 130 - Summer 2017 William Qiu 32

Creating User-defined Formats
To create and use your own formats,
1. Use the FORMAT procedure to create the format
2. Apply the format to specific variables by using a
FORMAT statement

PSTAT 130 - Summer 2017 William Qiu 33

Create User-defined Formats
We now learned to associated formats with variables. But sometimes we
might want to create custom formats for displaying variable values. For
example, we can format a product number so that it is displayed as
descriptive text.
With the FORMAT procedure, we can define our own formats for variables.
We can store our formats temporarily or permanently, and we can display a
list of all our formats and descriptions of their values.
General form of a PROC FORMAT step:
PROC FORMAT;
VALUE format-name range1=‘label’
range2=‘label’
. . . ;
RUN;

PSTAT 130 - Summer 2017 William Qiu 34

Creating User-defined Formats
Format-name
 Names the format you are creating (i.e., DATE7.)
 Cannot be more than 8 characters
 For character values, must have a dollar $ as the first character
, a letter or underscore as the second character, and no more
than 6 additional characters, number and underscores
 For numeric values, must have a letter or underscore as the
first character, and no more than 7 additional characters,
number and underscores
 Cannot end in a number
 Cannot be the same as a SAS format
 Does not end with a period in the VALUE statement

PSTAT 130 - Summer 2017 William Qiu 35

Creating User-define Formats
Labels
 Can be up to 32.767 characters in length
 Are typically text string enclosed in quotation marks,
although it is not required

Range(s): specifies one or more variable values and a

character string or an existing format.
 Can be single values or eg. 24, ‘S’…
 ranges of values, eg. 0-1500 (numeric), ‘A’-’M’ (characer)
 More examples next page

PSTAT 130 - Summer 2017 William Qiu 36

Creating User-defined Formats
Assigning labels to single numbers
Formatted
proc format;
value
value gender 1=‘Female’
2=‘Male’
Other=‘Miscoded’;
run;

Numeric Numeric
format data values Keyword
name
The statement begins with the keyword VALUE and ends
with a semicolon after all the labels have been assigned.

PSTAT 130 - Summer 2017 William Qiu 37

Creating User-defined Formats
Assigning labels to a range of numbers

proc format;
value boardfmt low-49=‘Below Average’
50-99=‘Average’
100-high=‘Above Average’;
run;

Numeric
data values Keywords

PSTAT 130 - Summer 2017 William Qiu 38

Creating User-defined Formats
Assigning labels to character and ranges of
character values
Character
format name

proc format;
value $grade ‘A’=‘Good’
‘B’-’D’=‘Fair’
‘F’=‘Poor’
‘I’,’U’=‘See Instructor’
‘F’=‘Poor’
Other=‘Miscoded’;
run;

Character Discrete Keyword

value range character values
PSTAT 130 - Summer 2017 William Qiu 39
Creating User-defined Formats
Step 1. Create the format
proc format;
value $codefmt ‘FLTAT’=‘Flight Attendant’
‘PILOT’=‘Pilot’;
run;

Step 2. Apply the format

proc print data=ia.empdata;
format Jobcode $codefmt.;
run;

PSTAT 130 - Summer 2017 William Qiu 40

Creating User-defined Formats
Step 1. Create the format
proc format;
value money low-<25000=‘Less than 25,000’
25000-50000=‘25,000 to 50,000’
50000-high=‘More than 50,000’;
run;

Step 2. Apply the format

proc print data=ia.empdata;
format Salary money.;
run;

PSTAT 130 - Summer 2017 William Qiu 41

Creating User-defined Formats
Define multiple formats within the same FORMAT
procedure
proc format;
value $codefmt ‘FLTAT’=‘Flight Attendant’
‘PILOT’=‘Pilot’;
value money low-<25000=‘Less than 25,000’
25000-50000=‘25,000 to 50,000’
50000-high=‘More than 50,000’;
run;

PSTAT 130 - Summer 2017 William Qiu 42

Applying User-defined Formats
proc print data=ia.empdata split=' ' noobs;
label LastName='Last Name'
FirstName='First Name'
Salary='Annual Salary';
format Salary money. Jobcode $codefmt.;
title1 'Salary Report in Categories';
run;

PSTAT 130 - Summer 2017 William Qiu 43

William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017 - William Qiu 2
Parking Lot Issues
 Homework is due in your registered section
 Data file location
 Default SAS directory
 List of formats – SAS Help
 SAS vs. User-defined formats

PSTAT 130 - Summer 2017 - William Qiu 3

Objectives
 Generate simple list reports using the PRINT procedure.
 Display Formatted Data Values
 Display selected variables (columns) in a list report.
 Display selected observations (rows) in a list report.
 Display a list report with column totals.

PSTAT 130 - Summer 2017 - William Qiu 4

PROC PRINT Default Output
 Use variable names as column headings
 Display all variables contained in data set
 Display all observations contained in data set
 Display variable values in their “native” format

PSTAT 130 - Summer 2017 - William Qiu 5

PROC PRINT – Key Options
With PROC PRINT options, you can display
 titles and footnotes
 descriptive column headings
 formatted data values
 specific variables and observations
 column totals and column subtotals
 page breaks for each subgroup

PSTAT 130 - Summer 2017 - William Qiu 6

PROC PRINT Syntax

PSTAT 130 - Summer 2017 - William Qiu 7

Defining Titles and Footnotes
 General form of the TITLE statement
TITLEn ‘text’

 General form of the FOOTNOTE statement

FOOTNOTEn ‘text’;

 Examples
 title1 ‘PSTAT 130 Homework #1’;
 footnote2 ‘Confidential’;

PSTAT 130 - Summer 2017 - William Qiu 8

Defining Titles and Footnote
Features of titles:
 Titles appear at the top of the page of output
 The default title is “The SAS System”
 You can have more than title line
 The value of n can be from 1 to 10 and refers to title line
 An unnumbered TITLE is equivalent to TITLE1
 Titles remain in effect until they are changed
 The null TITLE statement , title; cancels all titles

PSTAT 130 - Summer 2017 - William Qiu 9

Defining Titles and Footnote
Features of footnotes:
 Footnotes appear at the bottom of the page of output
 There is no default footnote
 You can have more than footnote
 The value of n can be from 1 to 10 and refers to footnote
line
 An unnumbered FOOTNOTE is equivalent to FOOTNOTE1
 Footnotes remain in effect until they are changed
 The null FOOTNOTE statement, footnote; cancels all
footnote
PSTAT 130 - Summer 2017 - William Qiu 10
Changing Titles and Footnotes
TITLEn or FOOTNOTEn
 replaces a previous title or footnote with the same
number
 Cancels all titles or footnotes with higher numbers

PSTAT 130 - Summer 2017 - William Qiu 11

Defining Titles and Footnotes
PROC PRINT Code Resultant Title(s)
proc print data=work.march;
title1 ‘The First Line’;
title2 ‘The Second Line’;
run;
proc print data=work.april;
title2 ‘The Next Line’;
run;
proc print data=work.may;
title ‘The Top Line’;
run;
proc print data=work.june;
title3 ‘The Third Line’;
run;
proc print data=work.july;
title;
run;

PSTAT 130 - Summer 2017 - William Qiu 12

Assigning Column Labels
 General form of the LABEL statement
LABEL variable=‘label’
variable=‘label’;
 Examples:
LABEL Units=‘Number of Units’
DOB=‘Date of Birth’;

PSTAT 130 - Summer 2017 - William Qiu 13

Assigning Column Labels
ia.empdata

Default Variable Names

PSTAT 130 - Summer 2017 - William Qiu 14

Assigning Column Labels
ia.empdata

PSTAT 130 - Summer 2017 - William Qiu 15

Assigning Column Labels
ia.empdata

PSTAT 130 - Summer 2017 - William Qiu 16

Using SAS Options
 General form of the OPTIONS statement:

OPTIONS option...;

 Run the following program to generate a list of all options

(the list will appear in the LOG window):

PROC OPTIONS;
RUN;

PSTAT 130 - Summer 2017 - William Qiu 17

Using SAS System Options
Selected SAS system options:
 DATE vs. NODATE
 Determines whether SAS displays the system date on output
 LINESIZE=width (LS=width)
 Determines size of output line before line break
 PAGESIZE=n (PS=n)
 Determines number of lines on page before page break
 NUMBER vs. NONUMBER
 Determines whether SAS displays page numbers
 PAGENO=n
 Determines starting page number
 Example:
Options nodate nonumber ls=72;
PSTAT 130 - Summer 2017 - William Qiu 18
Formatting Data Values

PSTAT 130 - Summer 2017 - William Qiu 19

Objectives
 Display formatted values using SAS formats in a list report
 Create user-defined formats using the FORMAT procedure
 Apply user-defined formats to variables in a list report

PSTAT 130 - Summer 2017 - William Qiu 20

Using SAS Formats
Enhance the readability of reports by formatting the data
values

PSTAT 130 - Summer 2017 - William Qiu 21

Using SAS Formats
Create custom formats to recode data values in a report

For a list of formats, see Formats, By Category in SAS Help

PSTAT 130 - Summer 2017 - William Qiu 22

Formatting Data Values

SAS
Format Report
Data
Set

PSTAT 130 - Summer 2017 - William Qiu 23

Formatting Data Values
 General form of the FORMAT statement
FORMAT variable(s) format;
 Example:
proc print data=ia.empdata split=' ';
format Salary dollar11.2;
run;

PSTAT 130 - Summer 2017 - William Qiu 24

What is a SAS Format?
<$> format<w>.<d>
Number of
Indicates a decimal places
character
format
Format name Required
delimiter

Total width (including decimal

places and special characters)

PSTAT 130 - Summer 2017 - William Qiu 25

SAS Formats
 Selected SAS formats

Format/ Result Description

Example
w.d standard numeric format
8.2 12234.21 Width=8, 2 decimal places
$w. standard character format
$5. KATHY width=5
COMMAw.d Commas in a number
COMMA9.2 12,234.21 Width=9, 2 decimal places
DOLLARw.d Dollar signs and commas in a number
DOLLAR10.2 $12,234.21 Width=10, 2 decimal places

PSTAT 130 - Summer 2017 - William Qiu 26

SAS Formats
Stored Value Format Displayed
Value
27134.2864 COMMA12.2 27,134.29
27134.2864 12.2 27134.29
27134.2864 DOLLAR12.2 $27,134.29
27134.2864 DOLLAR9.2 $27134.29
27134.2864 DOLLAR8.2 27134.29
27134.2864 DOLLAR5.2 27134
27134.2864 DOLLAR4.2 27E3

PSTAT 130 - Summer 2017 - William Qiu 27

Formatting Data Values
ia.empdata

PSTAT 130 - Summer 2017 - William Qiu 28

Date Formats

MMDDYYw. DATEw.
Format Displayed Format Displayed
Value Value
MMDDYY6. 101601 DATE7. 16OCT01
MMDDYY8. 10/16/01 DATE9. 16OCT2001
MMDDYY10. 10/16/2001

PSTAT 130 - Summer 2017 - William Qiu 29

DATE Formats and Serial Date Values

Stored Value Format Displayed Value

0 MMDDYY8. 01/01/60
0 MMDDYY10. 01/01/1960
1 DATE9. 02JAN1960
-1 WORDDATE. December 31, 1959
365 DDMMYY10. 31/12/1960
366 WEEKDATE. Sunday, January 1,
1961

PSTAT 130 - Summer 2017 - William Qiu 30

Creating User-defined Formats
To create and use your own formats,
1. Use the FORMAT procedure to create the format
2. Apply the format to specific variables by using a
FORMAT statement

PSTAT 130 - Summer 2017 - William Qiu 31

Create User-defined Formats
General form of a PROC FORMAT step:

PROC FORMAT;
VALUE format-name range1=‘label’
range2=‘label’
...;
RUN;

PSTAT 130 - Summer 2017 - William Qiu 32

Creating User-defined Formats
Format-name
 Names the format you are creating (i.e., DATE7.)
 Cannot be more than 8 characters
 For character values, must have a dollar $ as the first character
, a letter or underscore as the second character, and no more
than 6 additional characters, number and underscores
 For numeric values, must have a letter or underscore as the
first character, and no more than 7 additional characters,
number and underscores
 Cannot end in a number
 Cannot be the same as a SAS format
 Does not end with a period in the VALUE statement

PSTAT 130 - Summer 2017 - William Qiu 33

Creating User-define Formats
Labels
 Can be up to 32.767 characters in length
 Are typically enclosed in quotes, although it is not
required

Range(s)
 Can be single values or
 ranges of values

PSTAT 130 - Summer 2017 - William Qiu 34

Creating User-defined Formats
Assigning labels to single numbers
Formatted
proc format;
value
value gender 1=‘Female’
2=‘Male’
Other=‘Miscoded’;
run;

Numeric Numeric
format data values Keyword
name

PSTAT 130 - Summer 2017 - William Qiu 35

Creating User-defined Formats
Assigning labels to a range of numbers

proc format;
value boardfmt low-49=‘Below Average’
50-99=‘Average’
100-high=‘Above Average’;
run;

Numeric
data values Keywords

PSTAT 130 - Summer 2017 - William Qiu 36

Creating User-defined Formats
Assigning labels to character and ranges of
character values
Character
format name

proc format;
value $grade ‘A’=‘Good’
‘B’-’D’=‘Fair’
‘F’=‘Poor’
‘I’,’U’=‘See Instructor’
‘F’=‘Poor’
Other=‘Miscoded’;
run;

Character Discrete Keyword

value range character values
PSTAT 130 - Summer 2017 - William Qiu 37
Creating User-defined Formats
Step 1. Create the format

proc format;
value $codefmt “FLTAT”=“Flight Attendant”
“PILOT”=“Pilot”;
run;

Step 2. Apply the format

proc print data=ia.empdata;

format Jobcode $codefmt.;
run;

PSTAT 130 - Summer 2017 - William Qiu 38

Creating User-defined Formats
Step 1. Create the format

proc format;
value money low-25000=‘Less than 25,000’
25000-50000=‘25,000 to 50,000’
50000-high=‘More than 50,000’;
run;

Step 2. Apply the format

proc print data=ia.empdata;

format Salary money.;
run;

PSTAT 130 - Summer 2017 - William Qiu 39

Creating User-defined Formats
Define multiple formats within the same FORMAT
procedure

proc format;
value $codefmt ‘FLTAT’=‘Flight Attendant’
‘PILOT’=‘Pilot’;
value money low-25000=‘Less than 25,000’
25000-50000=‘25,000 to 50,000’
50000-high=‘More than 50,000’;
run;

PSTAT 130 - Summer 2017 - William Qiu 40

Applying User-defined Formats
proc print data=ia.empdata split=' ' noobs;
label LastName='Last Name'
FirstName='First Name'
Salary='Annual Salary';
format Salary money. Jobcode $codefmt.;
title1 'Salary Report in Categories';
run;

PSTAT 130 - Summer 2017 - William Qiu 41

Printing Selected Variables
 By default, PROC PRINT prints all variables in the data set
The VAR statement enables you to
 select variables to include in the report
 define the order of the variables in the report.
General form of the VAR statement:
VAR variable(s);

Select and order variables to print:

proc print data=ia.empdata;
var JobCode EmpID Salary;
run;
PSTAT 130 - Summer 2017 - William Qiu 42
Suppressing the Obs Column
The NOOBS option suppresses the row numbers on the left
side of the report.
General form of the NOOBS option:

PROC PRINT DATA=SAS-data-set NOOBS;

RUN;

Suppress the Obs column:

proc print data=ia.empdata noobs;
var JobCode EmpID Salary;
run;

PSTAT 130 - Summer 2017 - William Qiu 43

Subsetting Data: WHERE Statement
Produce a listing report that displays information for pilots
only.
The WHERE statement
 enables you to select observations that meet a certain
condition
 can be used with most SAS procedures.

PSTAT 130 - Summer 2017 - William Qiu 44

Subsetting Data: WHERE Statement
General form of the WHERE statement:
WHERE where-expression;
where-expression is a sequence of operands and operators.

Operands include
 variables or constants

Operators include
 comparison operators
 logical operators
 special operators
 functions
PSTAT 130 - Summer 2017 - William Qiu 45
Printing Selected Observations
Use the WHERE statement to control which observations
are processed.
proc print data=ia.empdata noobs;
var JobCode EmpID Salary;
where JobCode='PILOT';
run;

PSTAT 130 - Summer 2017 - William Qiu 46

WHERE Subsetting Example
 Print only employees who are Flight Attendants

proc print data=ia.empdata;

var JobCode EmpID Salary;
WHERE JobCode = ‘FLTAT’;
run;

 The WHERE statement selects employees whose JobCode

is “FLTAT”

PSTAT 130 - Summer 2017 - William Qiu 47

Comparison Operators
Mnemonic Symbol Definition Example
Equivalent
EQ = equal to where empnum eq 3374;
NE ^= not equal to where status ne fulltime;
~=
<>
GT > greater than where hiredate gt '01jun1982'd;

LT < less than where empnum < 2000;

GE >= greater than or equal to where empnum >= 3374;
LE <= less than or equal to where empnum <= 3374;
IN equal to one from a list of values where state in ('NC','TX');

• Character comparisons are case-sensitive.

• The IN operator allows commas or spaces to separate arguments.
PSTAT 130 - Summer 2017 - William Qiu 48
WHERE Examples
 Print only people under 40
 Where Age < 40;
 Where Age LT 40;
 Where Age <= 39;
 Where Age LE 39;
 Print only adults
 Where Age >= 18;
 Where Age GE 18;
 Where Age > 17;
 Where Age GT 17;
 Print only women
 Where sex = ‘F’;
 Where sex EQ ‘F’;
 Where sex <> ‘M’;
 Where sex NE ‘M’;

PSTAT 130 - Summer 2017 - William Qiu 49

Logical Operators
 AND (&)
 if both expressions are true, then the compound expression is true
where JobCode='FLTAT' and Salary>50000;
where JobCode='FLTAT' & Salary>50000;

 OR (|)
 if either expression is true, then the compound expression is true
where JobCode='PILOT' or JobCode='FLTAT';
where JobCode='PILOT' | JobCode='FLTAT';

 NOT
 can be combined with other operators to reverse the logic of a
comparison.
where not JobCode in('PILOT','FLTAT');

PSTAT 130 - Summer 2017 - William Qiu 50

Special Operators
Additional special operators supported by the WHERE
statement are
 BETWEEN-AND
 CONTAINS (?)
 LIKE
 sounds like
 IS MISSING (or IS NULL)

PSTAT 130 - Summer 2017 - William Qiu 51

Special Operators
Special operators include
 BETWEEN-AND
 selects observations in which the value of the variable falls within a
range of values, inclusively.
where Salary between 50000 and 70000;
Same as where 50000 <= Salary <= 70000;

 CONTAINS (?)
 selects observations that include the specified substring.
where LastName ? 'LAM';
(LAMBERT, BELLAMY, and ELAM are selected.)
Contrast In and ?

PSTAT 130 - Summer 2017 - William Qiu 52

Special Operators
The following are special operators :
 LIKE selects observations by comparing character values
to specified patterns.
 LIKE uses two “wildcard” symbols:
 A percent sign (%) replaces any number of characters.
 An underscore (_) replaces one character.

where Code like 'E_U%';

 Selects observations where the value of Code begins with

an E, followed by a single character, followed by a U,
followed by any number of characters.

PSTAT 130 - Summer 2017 - William Qiu 53

Special Operators
 The sounds like (=*) operator selects observations that
contain spelling variations of the word or words specified.
where Name=*'SMITH';

Selects names like SMYTHE or SMITT

 IS NULL or IS MISSING selects observations in which the
value of the variable is missing.
where Flight is missing;
where Flight is null;

 You can use the NOT logical operator to select non-

missing values.

PSTAT 130 - Summer 2017 - William Qiu 54

Adjusting Column Width
The WIDTH option controls the column width of each
variable on the PRINT report

General form of the WIDTH Option:

PROC PRINT DATA=SAS-data-set WIDTH=options;
RUN;

Example:
proc print data=ia.empdata width=uniform;
run;

PSTAT 130 - Summer 2017 - William Qiu 55

Adjusting Column Width
By default, PROC PRINT uses the formatted column width
for each variable. To set the column width to the same
size for each variable, use WIDTH=UNIFORM.

Note: Columns will not display smaller than the width of

their column headings.

PSTAT 130 - Summer 2017 - William Qiu 56

Requesting Column Totals
The SUM statement produces column totals for numeric
variables.
General form of the SUM statement:
SUM variable(s);

Example:
proc print data=ia.empdata noobs;
var JobCode EmpID Salary;
sum Salary;
run;
(Note: the SUM statement also produces subtotals if you
print the data in groups.)
PSTAT 130 - Summer 2017 - William Qiu 57
Sequencing and Grouping Observations

PSTAT 130 - Summer 2017 - William Qiu 58

Objectives
 Sequence (sort) observations in a SAS data set.
 Group observations in a list report.
 Print column subtotals in a list report.
 Control page breaks for subgroups

PSTAT 130 - Summer 2017 - William Qiu 59

Sorting a SAS Data Set
To request subgroup totals in PROC PRINT, the observations
in the data set must be grouped.
The SORT procedure
 rearranges the observations in a SAS data set
 can create a new SAS data set containing the rearranged
observations
 can sort on multiple variables
 can sort in ascending (default) or descending order
 does not generate printed output
 treats missing values as the smallest possible value.
PSTAT 130 - Summer 2017 - William Qiu 60
Sorting a SAS Data Set
General form of the PROC SORT step:
PROC SORT DATA=input-SAS-data-set
<OUT=output-SAS-data-set>;
BY <DESCENDING> by-variable(s);
RUN; CAUTION!
Overwrites the
Examples (sorts the data by Salary): original dataset
with the sorted
proc sort data=ia.empdata; data set
by Salary;
run;
Saves the
sorted data
proc sort data=ia.empdata out=work.jobsal; in a new
by Salary; data set
run;

PSTAT 130 - Summer 2017 - William Qiu 61

Sorting a SAS Data Set
When you include more than one Sort variable in the BY
statement:
 SAS sorts the data set by the first variable listed
 then sorts by the second variable WITHIN the values of
the first variable
 then sorts by the third variable WITHIN the values of the
second variable, etc

By default, SAS sorts in ascending order. The keyword

descending applies to the following variable.

PSTAT 130 - Summer 2017 - William Qiu 62

Sorting Exercise
 The admit dataset in Classes\Pstat130\data1 contains
hospital admitting information on patients
 Print the admit dataset
 Sort the admit dataset INTO A NEW dataset called
work.temp_admit
 Sort it by Sex Actlevel
 Print the new dataset
 Now sort by Actlevel Sex
 Print the new dataset

PSTAT 130 - Summer 2017 - William Qiu 63

Create listings for subgroups
The BY statement enables you to
 Create a separation between listings of several subgroups
 Identify each subgroup of the top of its listing
 Data set must be sorted on BY variable first

General form of the BY statement:

BY variable(s);
Example:

proc print data=data1.diabetes;

BY Sex;
run;
PSTAT 130 - Summer 2017 - William Qiu 64
 The SAS System 18:44 Tuesday, October 4, 2011 26

--------------------------------------------- Sex=F ----------------------------------------------

Act
Obs ID Name Age Date Height Weight Level Fee

1 2462 Almers, C 34 3 66 152 HIGH 124.80

2 2571 Nunnelly, A 44 19 66 140 HIGH 149.75
3 2575 Quigley, M 40 8 69 163 HIGH 124.80
4 2589 Wilcox, E 41 16 67 141 HIGH 149.75
5 2501 Bonaventure, T 31 17 61 123 LOW 149.75
6 2568 Eberhardt, S 49 27 64 172 LOW 124.80
7 2572 Oberon, M 28 17 62 118 LOW 85.20
8 2588 Ivan, H 22 20 63 139 LOW 85.20
9 2523 Johnson, R 43 31 63 137 MOD 149.75
10 2552 Reberson, P 32 9 67 151 MOD 149.75
11 2584 Takahashi, Y 43 29 65 123 MOD 124.80

--------------------------------------------- Sex=M ----------------------------------------------

Act
Obs ID Name Age Date Height Weight Level Fee

12 2458 Murray, W 27 1 72 168 HIGH 85.20

13 2544 Jones, M 29 6 76 193 HIGH 124.80
14 2586 Derber, B 25 23 75 188 HIGH 85.20
15 2539 LaMance, K 51 4 71 158 LOW 124.80
16 2563 Pitts, D 34 22 73 154 LOW 124.80
17 2579 Underwood, K 60 22 71 191 LOW 149.75
18 2555 King, E 35 13 70 173 MOD 149.75
19 2574 Peterson, V 30 6 69 147 MOD 149.75
20 2578 Cameron, L 47 5 72 173 MOD 124.80
21 2595 Warren, C 54 7 71 183 MOD 149.75

PSTAT 130 - Summer 2017 - William Qiu 65

Printing Subtotals and Grand Totals
Print the data set grouped by JobCode with a subtotal for the
Salary column for each JobCode.
proc sort data=ia.empdata
out=work.empdata;
by JobCode;
run;
proc print data=work.empdata;
by JobCode;
sum Salary;
run;
Using a BY statement and a SUM statement together in
a PROC PRINT step produces subtotals and grand
totals.
PSTAT 130 - Summer 2017 - William Qiu 66
------------------------------------------ ActLevel=LOW ------------------------------------------

Obs ID Name Sex Age Date Height Weight Fee

8 2501 Bonaventure, T F 31 17 61 123 149.75

9 2568 Eberhardt, S F 49 27 64 172 124.80
10 2572 Oberon, M F 28 17 62 118 85.20
11 2588 Ivan, H F 22 20 63 139 85.20
12 2539 LaMance, K M 51 4 71 158 124.80
13 2563 Pitts, D M 34 22 73 154 124.80
14 2579 Underwood, K M 60 22 71 191 149.75
-------- ------
ActLevel 844.30

------------------------------------------ ActLevel=MOD ------------------------------------------

Obs ID Name Sex Age Date Height Weight Fee

15 2523 Johnson, R F 43 31 63 137 149.75

16 2552 Reberson, P F 32 9 67 151 149.75
17 2584 Takahashi, Y F 43 29 65 123 124.80
18 2555 King, E M 35 13 70 173 149.75
19 2574 Peterson, V M 30 6 69 147 149.75
20 2578 Cameron, L M 47 5 72 173 124.80
21 2595 Warren, C M 54 7 71 183 149.75
-------- ------
ActLevel 998.35
======
2687.0

PSTAT 130 - Summer 2017 - William Qiu 67

Page Breaks
Use the PAGEBY statement to put each subgroup on a
separate page.
General form of the PAGEBY statement:
PAGEBY by-variable;
Example:
proc print data=work.empdata;
by JobCode;
pageby JobCode;
sum Salary;
run;

The PAGEBY statement must be used with a BY statement.

PSTAT 130 - Summer 2017 - William Qiu 68
Printing Sorted Observations
If you attempt to use the BY statement on an un-sorted
data set, you will get an error.
ERROR: Data set IA.EMPDATA is not sorted in ascending
sequence. The current by-group has JobCode = PILOT
and the next by-group has JobCode = FLTAT.
NOTE: The SAS System stopped processing this step
because of errors.
NOTE: There were 2 observations read from the data
set IA.EMPDATA.

PSTAT 130 - Summer 2017 - William Qiu 69

Identifying Observations

PSTAT 130 - Summer 2017 - William Qiu 70

Objectives
 Use the ID statement to identify observations.
 Combine the BY and ID statements to produce special
formatting.

PSTAT 130 - Summer 2017 - William Qiu 71

Identifying Observations
The ID statement enables you to
 suppress the Obs column in the report
 specify which variable(s) should replace the Obs column.

General form of the ID statement:

ID variable(s);

Example:
proc print data=ia.empdata;
id JobCode;
var EmpID Salary;
run;

PSTAT 130 - Summer 2017 - William Qiu 72

Special BY-Group Formatting
When the ID and BY statements specify the same variable,
 the Obs column is suppressed
 the BY line is suppressed
 the ID/BY variable prints in the leftmost column
 each ID/BY value only prints at the start of each BY group
(and on the subtotal line, if a SUM statement is used).

PSTAT 130 - Summer 2017 - William Qiu 73

Special BY-Group Formatting
 Specify JobCode in the BY and ID statements to change
the report format.
proc sort data=ia.empdata out=work.empdata;
by JobCode;
run;
proc print data=work.empdata;
by JobCode;
id JobCode;
sum Salary;
run;

PSTAT 130 - Summer 2017 - William Qiu 74

Special BY-Group Formatting
The SAS System

Job Emp
Code ID LastName FirstName Salary

FLTAT 0040 WILLIAMS ARLENE M. 23666.12

0071 PERRY ROBERT A. 21957.71
0091 SCOTT HARVEY F. 32278.40
0106 THACKER DAVID S. 24161.14
----- ---------
FLTAT 102063.37

PILOT 0031 GOLDENBERG DESIREE 50221.62

0082 MCGWIER-WATTS CHRISTINA 96387.39
0355 BELL THOMAS B. 59803.16
0366 GLENN MARTHA S. 120202.38
----- ---------
PILOT 326614.55
=========
428677.92
PSTAT 130 - Summer 2017 - William Qiu 75
William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017 - William Qiu
Objectives
 State the components of a SAS program
 State the modes in which you can run SAS

PSTAT 130 - Summer 2017 - William Qiu

SAS Programs
 A SAS Program is a collection of statements, grouped into
a sequence of steps, that the programmer submits for
execution
DATA steps are typically used to create SAS
Raw data sets
Data
DATA SAS PROC Report
Step Data Step
set
SAS
Data PROC steps are typically used to process
set SAS data sets (i.e., generate reports and
graphs, sort and modify data)

PSTAT 130 - Summer 2017 - William Qiu

SAS Programs
data work.staff;
infile 'X:\PStat 130\data1\emplist.dat'; DATA Step
input LastName $ 1-20 FirstName $ 21-30
JobTitle $ 36-43 Salary 54-59;
run;

proc print data=work.staff;

run;
PROC Steps
proc means data=work.staff;
class JobTitle;
var Salary;
run;

PSTAT 130 - Summer 2017 - William Qiu

Step Boundaries
SAS steps begin with a
 DATA statement
 PROC statement

SAS detects the end of a step when it encounters

 A RUN statement (for most steps)
 A QUIT statement (for some procedures)
 The beginning of another step (DATA statement or PROC
statement)

PSTAT 130 - Summer 2017 - William Qiu

Step Boundaries
data work.staff;
infile 'raw-data-file';
input LastName $ 1-20 FirstName $ 21-30
JobTitle $ 36-43 Salary 54-59;
run;

proc print data=work.staff;

run;

proc means data=work.staff;

class JobTitle;
var Salary;
run;

PSTAT 130 - Summer 2017 - William Qiu

Running a SAS Program
You can invoke SAS in
 Interactive windowing mode (SAS windowing
environment)
 Interactive menu-driven mode (Enterprise Guide,
SAS/ASSIST , SAS/AF, or SAS/EIS software)
 Batch mode
 Noninteractive mode

PSTAT 130 - Summer 2017 - William Qiu

Fundamental Concepts

PSTAT 130 - Summer 2017 - William Qiu

Objectives
 Define the components of a SAS data set
 Define a SAS variable
 Identify a missing value and a SAS date value
 State the naming conventions for SAS data sets and
variables
 Explain SAS syntax rules
 Investigate a SAS data set using the CONTENTS and PRINT
procedures

PSTAT 130 - Summer 2017 - William Qiu

SAS Data Sets
Other
External
Data Entry Software
File
Files

Conversion
Process

SAS Data
Set
Descriptor Data
Portion Portion
PSTAT 130 - Summer 2017 - William Qiu
SAS Data Sets
SAS data sets have a descriptor portion and a data portion.
Descriptor General data set information
Portion * data set name * data set label
* date/time created * storage information
* number of observations

Information for each variable

* Name * Type * Length * Position
* Format * Informat * Label

Data Portion Observations for each variable

Name Age Height Weight
John 19 69 180
Mary 22 63 130
John 21 67 165
PSTAT 130 - Summer 2017 - William Qiu
Browsing the Descriptor Portion
The descriptor portion of a SAS data set contains
 general information about the SAS data set (such as
 data set name and number of observations)
 variable attributes (name, type, length, position,
informat, format, label).
The CONTENTS procedure (PROC CONTENTS) displays the
descriptor portion of a SAS data set.

PSTAT 130 - Summer 2017 - William Qiu

Browsing the Descriptor Portion
General form of the CONTENTS procedure:

PROC CONTENTS DATA=<SAS-data-set>;

RUN;

Example:
proc contents data=data1.empdata;
run;

PSTAT 130 - Summer 2017 - William Qiu

Sample CONTENTS Output
The SAS System

The CONTENTS Procedure

Data Set Name WORK.EMPDATA Observations 8

Member Type DATA Variables 5
Engine V9 Indexes 0
Created Sunday, October 03, Observation Length 48
2010 04:12:45 PM
Last Modified Sunday, October 03, Deleted Observations 0
2010 04:12:45 PM
Protection Compressed NO
Data Set Type Sorted NO
Label
Data Representation WINDOWS_32
Encoding wlatin1 Western (Windows)

Alphabetic List of Variables and Attributes

# Variable Type Len

1 EmpID Char 4
3 FirstName Char 13
4 JobCode Char 5
2 LastName Char 13
5 Salary Num 8

PSTAT 130 - Summer 2017 - William Qiu

SAS Data Sets: Data Portion
The data portion of a SAS data set is a rectangular table of
character and/or numeric data values.
Job
LastName FirstName Code Salary
Variable
Names
GOLDENBERG DESIREE PILOT 50221.62
WILLIAMS ARLENE M. FLTAT 23666.12
PERRY ROBERT A. FLTAT 21957.71
MCGWIER-WATTS CHRISTINA PILOT 96387.39
Variable
SCOTT HARVEY F. FLTAT 32278.40 Values
THACKER DAVID S. FLTAT 24161.14
BELL THOMAS B. PILOT 59803.16
GLENN MARTHA S. PILOT 120202.38

Character Values Numeric Values

PSTAT 130 - Summer 2017 - William Qiu

SAS Variable Values
There are two types of variables:
Character
 contains any value: letters, numbers, special characters,
and blanks. Character values are stored with a length of 1
to 32,767 bytes. One byte equals one character.
Numeric
 stored as floating point numbers in 8 bytes of storage by
default. Eight bytes of floating point storage provide
space for 16 or 17 significant digits. You are not restricted
to 8 digits.

PSTAT 130 - Summer 2017 - William Qiu

SAS Data Set and Variable Names
SAS names:
 can be 32 characters long.
 can be uppercase, lowercase, or mixed-case.
 must start with a letter or underscore. Subsequent
characters can be letters, underscores, or numeric digits.

PSTAT 130 - Summer 2017 - William Qiu

Valid SAS Names
Which of the following are valid SAS names?
A. data5mon
B. 5monthsdata
C. five months data
D. five_month_sdata
E. data#5

PSTAT 130 - Summer 2017 - William Qiu

SAS Date Values
 SAS stores date values as numeric values.
 A SAS date value is stored as the number of days between
January 1, 1960, and a specific date.

PSTAT 130 - Summer 2017 - William Qiu

Missing Data Values
A value must exist for every variable for each observation.
Missing values are valid values.
Job
LastName FirstName Code Salary

GOLDENBERG DESIREE PILOT 50221.62

WILLIAMS ARLENE M. FLTAT 23666.12
PERRY ROBERT A. FLTAT .
MCGWIER-WATTS CHRISTINA PILOT 96387.39
SCOTT HARVEY F. 32278.40

A character missing A numeric

value is displayed missing value
as a blank. is displayed as
a period.

PSTAT 130 - Summer 2017 - William Qiu

Browsing the Data Portion
The PRINT procedure displays the data portion of a SAS
data set.
By default, PROC PRINT displays
 all variables
 all observations
 an Obs column on the left side.

PSTAT 130 - Summer 2017 - William Qiu

Browsing the Data Portion
General form of the PRINT procedure:

PROC PRINT DATA=<SAS-data-set>;

RUN;

Example:

proc print data=work.empdata;

run;

PSTAT 130 - Summer 2017 - William Qiu

SAS Data Set Terminology
 SAS documentation and text in the SAS windowing
environment use the following terms interchangeably:
 SAS Data Set = SAS Table
 Variable=Column
 Observation=Row

PSTAT 130 - Summer 2017 - William Qiu

SAS Syntax Rules
 SAS statements usually begin with a keyword and always
end with a semi-colon.
 SAS statements are free-format.
 One or more blanks or special characters can be used to
separate words.
 Statements can begin and end in any column.
 A single statement can span multiple lines.
 Several statements can be on the same line.

PSTAT 130 - Summer 2017 - William Qiu

Unconventional Spacing, Poor Readability

libname mydrive ‘W:\Classes\Pstat

130\data1'; options nonumber nodate ls=80;
data staff; set mydrive.empdata; keep
lastname firstname jobcode salary; run;
proc contents data=work.staff; run; proc
print noobs; run;

PSTAT 130 - Summer 2017 - William Qiu

Conventional Spacing, Better Readability
libname mydrive 'E:\UCSB\Pstat 130\data1';
options nonumber nodate ls=80;

data staff;
set mydrive.empdata;
keep lastname firstname jobcode salary;
run;

proc contents data=work.staff;

run;

proc print noobs;

run;

PSTAT 130 - Summer 2017 - William Qiu

SAS Libraries - Details
A SAS library is a collection of SAS data files
 Use the LIBNAME statement to assign a “libref” (library
reference” to a SAS data library
General form:
LIBNAME libref 'SAS-data-library‘ <options>;

 Rules for Naming a Libref :

 must be 8 characters or less
 must begin with a letter or underscore
 remaining characters are letters, numbers, or underscores.

PSTAT 130 - Summer 2017 - William Qiu

Two-level SAS Data Filenames
Every SAS data file has a two-level name:

General Form
Libref.filename

Examples:
 ia.empdata
 work.staff

PSTAT 130 - Summer 2017 - William Qiu

Browsing SAS Libraries
During an interactive SAS session, the LIBNAME window enables you to
investigate the contents of a SAS data library.

In the LIBNAME window, you can

 view a list of all the libraries available during your current SAS session
 drill down to see all members of a specific library
 display the descriptor portion of a SAS data set

PSTAT 130 - Summer 2017 - William Qiu

Browsing a SAS Data Library
Use the _ALL_ keyword to list all the SAS files in the library
and the NODS option to suppress the descriptor portions of
the data sets.
 General form of the NODS option:

PROC CONTENTS DATA=libref._ALL_ NODS;

RUN;

 NODS must be used in conjunction with the keyword

_ALL_.
proc contents data=ia._all_ nods;
run;

PSTAT 130 - Summer 2017 - William Qiu

Browsing a SAS Library – Partial Output
The SAS System

The CONTENTS Procedure

Directory

Libref IA
Engine V9
Physical Name E:\UCSB\pstat 130\data1
File Name E:\UCSB\pstat 130\data1

Member File
# Name Type Size Last Modified

1 ADMIT DATA 9216 12Apr04:10:16:08

2 ADMITJUNE DATA 9216 12Apr04:10:16:08
3 ALLGOALS DATA 5120 12Apr04:10:16:08
4 ALLGOALS2 DATA 5120 12Apr04:10:16:08
5 ALLSALES DATA 5120 12Apr04:10:16:08
6 ALLSALES2 DATA 5120 12Apr04:10:16:08
7 APRTARGET DATA 17408 12Apr04:10:16:08
8 CHICAGO DATA 17408 12Apr04:10:16:14

PSTAT 130 - Summer 2017 - William Qiu

William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017- William Qiu
Objectives
 Identify the five sources of SAS data
 Create temporary and permanent SAS data sets
 Use list, column and formatted input to read raw data
 Explain how the DATA step processes data

PSTAT 130 -Summer 2017- William Qiu

Fundamental Principles
 The source, structure and format of your data determines
the methods used to bring it into SAS
 Source – where the data are stored (e.g., paper, text file,
another program file)
 Structure – what variables are present in the data (e.g., last
name, hire date, salary, flight number, destination)
 Format – how are values displayed (e.g., in neat columns,
separated by spaces, 10/12/2010 vs. 12OCT2010)

PSTAT 130 - Summer 2017- William Qiu

Five Data Sources
Existing SAS Other Raw Data in Raw Data in
Data Entry
data set Software SAS Program External File

SAS DATA step

New SAS
data set

PSTAT 130 - Summer 2017- William Qiu

Creating a SAS data set
Data Source KEYWORD/Method
Data in a permanent SAS dataset or SET
created by another SAS program
Data created in another software Import (interactive)
program (e.g., Excel)
Data Entry Table Editor
Raw Data in a SAS Program DATALINES
Raw Data in an external text file INFILE

PSTAT 130 - Summer 2017- William Qiu

Using an Existing SAS Data Set
If your data are in an existing SAS data set:
 Use the SET statement to read the existing SAS data set.
The data step creates a new SAS data set.
Example:
data work.temployees;
set ia.empdata;
run;

SET can refer to any existing SAS data set, temporary or

permanent
PSTAT 130 - Summer 2017- William Qiu
Importing Data from Other Software
If your data were created by other software (e.g., Excel)
 Use the IMPORT option on the File menu to read it in and
create a new SAS data set
 You will need to tell SAS
 what format the data are in
 where the file is stored
 what character is used to separate data values, (i.e., the
“delimiter “)
 Save your SAS data set in an existing library or create a
new library

PSTAT 130 - Summer 2017- William Qiu

Data Entry in SAS Table Editor
 Start SAS
 Go to the Tools Menu and Open the Table Editor
 Enter data values into the table
 Set the name, type, and width for each variable/column
using Data Menu - Column Attributes (or right-click on
each column heading)
 Save the data set into a library (you can create a library at
this time, if desired)
 You can now use this SAS data set in your program

PSTAT 130 - Summer 2017- William Qiu

Create SAS Data Set from Raw Data
To create a SAS data set from raw data you need to:
1. Use a DATA statement to name the SAS data set
2. Use an INFILE or DATALINES statement to identify where
to find the raw data
3. Use an INPUT statement to identify the variables in the
raw data file

PSTAT 130 - Summer 2017- William Qiu

Naming a SAS Data Set
General form of the DATA statement;
 This DATA statement creates a temporary SAS data set
data work.dfwlax;
set ia.dfwlax;
run;

 This DATA statement creates a permanent SAS data set

libname mydrive 'D:\';

data mydrive.dfwlax;
set ia.dfwlax;
run;
PSTAT 130 - Summer 2017- William Qiu
Reading Raw Data within a SAS Program
If the raw data are contained in the SAS Program:
 Use the DATALINES keyword, followed by the raw data lines.
Example:
data work.sample;
input firstname $ gender $ age;
datalines;
John Male 22
Jane Female 19
;
run;

This method works well for small amounts of data.

PSTAT 130 - Summer 2017- William Qiu
Pointing to a Raw Data File
If the raw data are contained in a separate text file:
 Use the INFILE statement, followed by a file specification.
Example:
data work.sample;
infile 'D:\UCSB\sample.txt';
input name $ gender $ age;
run;

 This method is much more common than DATALINES.

 The file specification does NOT use a library reference.
 The INFILE statement comes before the INPUT statement.

PSTAT 130 - Summer 2017- William Qiu

DATALINES versus INFILE
 The DATALINES method:
 Allows you to see your data directly
 Is good for small amounts of data
 Is often used to create “test” data sets
 The INFILE method
 Is more common
 Is necessary if your data comes from an outside source (i.e.,
client, download, website, etc)
 Is preferred for large data sets
 Allows you to easily re-run your programs on updated data

PSTAT 130 - Summer 2017- William Qiu

Reading Data Fields
The INPUT statement tells SAS the name and type of each
variable in the data set, and how to read the data.
Simple form of the INPUT statement

INPUT variable <$> <options>

General form of the INPUT statement:

INPUT <pointer-control> variable <$> <&> <@ | @@>;

INPUT variable <$> start-column <-- end-column> <.decimals> <@ | @@>;

INPUT <pointer-control> variable <:|&|~> <informat.> <@ | @@>;

PSTAT 130 - Summer 2017- William Qiu

INPUT Statement - Example
data work.students;
input firstname $ gender $ age;
datalines;
John Male 19
Wendy Female 22
;
run;

PSTAT 130 - Summer 2017- William Qiu

Types of Raw Data Input
 List Input - each data value is separated by a space as a
“delimiter”
 John Male 22
 Wendy Female 19
 Column Input – each data value is in a fixed location
 John Male 22
 Wendy Female 19
 Formatted Input – uses SAS formats (called informats)
 John Male 4/12/91
 Elizabeth Female 8/24/90

PSTAT 130 - Summer 2017- William Qiu

Simple List Input
If your data values are separated by a single space, use list
input
Example:
data work.students;
input Name $ Team $ Age;
datalines;
David Male 19
Amelia Female 23
Ravi Male 17
Ashley Female 20
Jim Male 26
;
run;

PSTAT 130 - Summer 2017- William Qiu

Column Input
If your data values are in fixed columns, and consist of
“standard” character and numeric values, use column
input
Example:
data work.students;
input Name $ 1-6 Gender $ 9-14 Age 18-20;
datalines;
David Male 19
Amelia Female 23
Ravi Male 17
Ashley Female 20
Jim Male 26
;
run;
PSTAT 130 - Summer 2017- William Qiu
Formatted Input
 If your data contains “non-standard” values, use
formatted input, called “informats”
informat

data students;
input Name $ Gender $ Age Enroll mmddyy8.;
datalines;
David Male 19 06/18/10
non-standard data
Amelia Female 23 08/02/10
(dates)
Ravi Male 17 07/22/10
Ashley Female . 09/14/10
Jim Male 26 08/26/10
;
run;

PSTAT 130 - Summer 2017- William Qiu

Format versus Informat
 A Format controls the way data look when you print
them.
 An Informat controls the way SAS reads data in.

PSTAT 130 - Summer 2017- William Qiu

William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017 - William Qiu
Review – Reading Data into SAS
There are three steps to reading raw data
1. DATA statement
2. INFILE or DATALINES statement
3. INPUT statement

There are three basic types of input

1. List input
Input Name $ GPA Score
2. Column input
Input Name $ 1-8 GPA 10-12 Score 14-16
3. Formatted input
Input Name $ 8. GPA 3.1 Score 3.
PSTAT 130 - Summer 2017 - William Qiu
Class Exercise
The data set below contains the 2010 population estimates
(in millions) of several U.S. states. Which type of input
statement would be used for the data set below?
Calif 36.9
Texas 24.8
NewYork 19.5
Florida 18.5
Illinois 12.9

Write the Input statement.

PSTAT 130 - Summer 2017 - William Qiu

Attributes of List Input
 All data values must be separate by a single space
 All variables must be in standard format
 Character and numeric values cannot contain spaces
 Character values cannot be longer than 8 characters
 Numeric values cannot contain commas or dollar signs
 Dates will be read as characters rather than date values

PSTAT 130 - Summer 2017 - William Qiu

Class Exercise (con’t)
The data set below contains the 2010 population estimates
(in millions) of several U.S. states. Which type of input
statement would be used for the data set below?
California 36.9
Texas 24.8
New York 19.5
Florida 18.5
Illinois 12.9

Write the Input statement.

PSTAT 130 - Summer 2017 - William Qiu

Attributes of Column Input
 The data values must occupy the same columns within
each observation
 This is called “fixed” or “aligned”
 Character variables can
 be longer than 8 characters
 contain spaces
 You can skip some data fields, if desired
 The data must be in “standard” format
 Numbers may not contain commas or dollar signs
 Dates will be read as character, instead of numeric,
variables

PSTAT 130 - Summer 2017 - William Qiu

Class Exercise (con’t)
The data set below contains the 2010 population estimates
of several U.S. states. Which type of input statement
would be used for the data set below?
California 36,961,664
Texas 24,782,302
New York 19,541,453
Florida 18,537,969
Illinois 12,910,409

Write the Input statement.

PSTAT 130 - Summer 2017 - William Qiu

Attributes of Formatted Input
 Data can be in “non-standard” format
 Numbers can contain commas and dollar signs
 Dates can be read as numeric variables
 Data can be free-form or fixed text files

PSTAT 130 - Summer 2017 - William Qiu

Reading Excel Spreadsheets
 Create a SAS data set from an Excel spreadsheet using the
Import Wizard.
 Create a SAS data set from an Excel spreadsheet using
PROC IMPORT.

PSTAT 130 - Summer 2017 - William Qiu

The IMPORT Procedure
General form of the IMPORT procedure

PROC IMPORT OUT=SAS-data-set

DATAFILE='external-file-name‘
DBMS=file-type;
GETNAMES=YES;
RUN;

PSTAT 130 - Summer 2017 - William Qiu

 Excel Import Example
PROC IMPORT OUT= WORK.tdfwlax
DATAFILE= “D:\Pstat 130\data1\DallasLA.xls”
DBMS=XLS REPLACE;
SHEET="DFWLAX";
GETNAMES=YES;
MIXED=NO;
SCANTEXT=YES;
USEDATE=YES;
SCANTIME=YES;
RUN;

PSTAT 130 - Summer 2017 - William Qiu

Assigning Variable Attributes
SAS allows you to:
 Assign permanent attributes to SAS variables.
 Change or override permanent variable attributes.

PSTAT 130 - Summer 2017 - William Qiu

Default Variable Attributes
When a variable is created in a DATA step, the
 name, type, and length of the variable are automatically
assigned
 remaining attributes such as label and format are not
automatically assigned.
When the variable is used in a later PROC step, the output
uses:
 the variable name
 a system-determined format.

PSTAT 130 - Summer 2017 - William Qiu

Specifying Variable Attributes
Use LABEL and FORMAT statements in the
 DATA step to permanently assign the attributes (stored in
the data set descriptor portion).
 PROC step to temporarily assign the attributes (for the
duration of the step only)

PSTAT 130 - Summer 2017 - William Qiu

 Comparison of assignment in DATA
and PROC steps
DATA work.bonus;
Set pstatlib.empdata;
Bonus = Salary * .1;
Label Bonus = ‘Annual Bonus’;
Format Bonus Dollar12.2; Permanent Attributes
Run;

PROC PRINT data=work.bonus;

Label Bonus = ‘Incentive Bonus’; Tempoarary Attributes
Format Bonus Dollar12.;
Run;

PSTAT 130 - Summer 2017 - William Qiu

The DATASETS Procedure
You can use the DATASETS procedure to modify a variable’s
 name
 label
 format
 informat.

PSTAT 130 - Summer 2017 - William Qiu

The DATASETS Procedure
General form of PROC DATASETS for changing variable
attributes:
PROC DATASETS LIBRARY=libref ;
MODIFY SAS-data-set ;
RENAME old-name-1=new-name-1
<. . . old-name-n=new-name-n>;
LABEL variable-1='label-1'
<. . . variable-n='label-n'>;
FORMAT variable-list-1 format-1
<. . . variable-list-n format-n>;
INFORMAT variable-list-1 informat-1
<. . . variable-list-n informat-n>;
RUN;
PSTAT 130 - Summer 2017 - William Qiu
Selecting Variables
Use the KEEP or DROP statement to select variables from
the data set.
 The KEEP statement identifies the variables that will
remain in the data set. Only the variables in the KEEP
statement will remain. All others are eliminated.
 The DROP statement identifies variables to be eliminated
from the data set. All variables except those in the DROP
statement will remain.

PSTAT 130 - Summer 2017 - William Qiu

KEEP vs KEEP= and DROP vs. DROP=
Use the KEEP or DROP statement to eliminate variables from the output data
set. These variables can still be used in SAS expressions.

Data libref.new-data-set;
KEEP variables;
or
DROP variables;
run;

Use the KEEP= or DROP= option in a SET statement to eliminate variables

from the input statement. The eliminated variables cannot be used in
expressions.
DATA libref.new-data-set;
SET SAS-data-set(KEEP=variables)
or
SET SAS-data-set(DROP=variables)
run;

PSTAT 130 - Summer 2017 - William Qiu

DROP & KEEP Examples
 Use DROP if you want to keep most of the variables
data work.tempdata;
Set ia.empdata;
Drop EmpID Hire;
run;

 Use KEEP if you only want to keep a few variables

data work.tempdata;
Set ia.empdata;
Keep Firstname Lastname;
run;

PSTAT 130 - Summer 2017 - William Qiu

Creating New Variables
Use variable assignment statements in the DATA step to
create new variables
An assignment statement:
 evaluates an expression
 assigns the resulting value to a variable.
General form of an assignment statement
DATA output-SAS-data-set;
SET input-SAS-data-set;
variable = expression;
RUN;

PSTAT 130 - Summer 2017 - William Qiu

SAS Expressions
An expression contains operands and operators that form a
set of instructions that produce a value.

Operands are
 Variable names
 Constants
Operators are
 Arithmetic symbols (+, -, /, *, etc)
 SAS functions

PSTAT 130 - Summer 2017 - William Qiu

Using Operators
Selected operators for basic arithmetic calculations in an
assignment statement:
Operator Action Example Priority
+ Addition Sum=x+y; III
- Subtraction Diff=x-y; III
* Multiplication Product=x*y; II
/ Division Divide=x/y; II
** Exponent Raise=x**y; I
- Negative Negative = -x; I

PSTAT 130 - Summer 2017 - William Qiu

Variable Assignment
Here are some examples of creating new variables using
arithmetic operators:
 TotalComp = Salary + Bonus;
 NetPay = GrossPay – Tax
 NewPay = Salary * (1 + Raise);
 Percent = Score/Maximum

PSTAT 130 - Summer 2017 - William Qiu

Using SAS Functions
A SAS function is a routine that returns a value that is
determined from specified arguments.

General form of a SAS function:

Function-name(argument1, argument2, …);

Example:

Total=sum(Salary, Bonus);

PSTAT 130 - Summer 2017 - William Qiu

Using SAS Functions
SAS functions
 perform arithmetic operations
 compute sample statistics (for example: sum, mean, and
standard deviation)
 manipulate SAS dates and process character values
 perform many other tasks.

Sample statistics functions ignore missing values.

PSTAT 130 - Summer 2017 - William Qiu

Numeric Functions
 MIN(x,y,z)- Returns the smallest value
 MAX(x,y,z)- Returns the largest value
 MEDIAN(x,y,z)- Calculates the median
 MEAN(x,y,z) – Calculates the average value
 STD(x,y,z) – Returns the standard deviation of the values
 ABS(x,y,z) – Returns the absolute value
 FACT(x) – Calculates the factorial, x!
 COS(x) – Returns the cosign of x
 TAN(x) – Returns the tangent of x

PSTAT 130 - Summer 2017 - William Qiu

Character Functions
 LOWCASE(x) – Converts all letters in x to lower case
 UPCASE(x) – Converts all letters in x to upper case
 TRIM(x) – Removes all trailing blanks
 SUBSTR(string,position,length) – Returns a subset of
given length within the characters in the string, starting at
the given position:
 SUBSTR(‘Statistics’,1,4) = ‘Stat’
 REVERSE(string) – reverse the order of letters
 REVERSE(‘STATISTICS’) = ‘SCITSITATS’

PSTAT 130 - Summer 2017 - William Qiu

Date Functions
TODAY() obtains the date value from the system clock.
MDY(month,day,year) uses numeric month, day, and year values
to return the corresponding SAS date value.
YEAR(SAS-date) extracts the year from a SAS date and returns a
four-digit value for year.
QTR(SAS-date) extracts the quarter from a SAS date and returns
a number from 1 to 4.
MONTH(SAS-date) extracts the month from a SAS date and
returns a number from 1 to 12.
WEEKDAY(SAS-date) extracts the day of the week from a SAS
date and returns a number from 1 to 7, where 1 represents
Sunday, and so on.

PSTAT 130 - Summer 2017 - William Qiu

Using SAS Date Constants
Use a Date Constant to return a SAS date value for a specific
date.

Example:
evaldate = ’14FEB2009’d;

sets the value of evaldate to 17942, which is the SAS date

value (number of days since 1960) corresponding to Feb 14,
2009. The text portion can be in the form of ‘ddmmmyyyy’ or
‘ddmmmyy’ .

PSTAT 130 - Summer 2017 - William Qiu

DateTime Values and DATEPART
A SAS datetime value is interpreted as the number of
seconds between midnight, January 1, 1960, and a
specific date and time.
Example: 12/01/2000 9:15 am is stored as 1,291,281,300
seconds since 01/01/1960 12:00 am.

The DATEPART function will return just the Date portion,

14945, which is be the number of days since 01/01/1960.
Example:
Birthdate = datepart(Birthdatetime);

PSTAT 130 - Summer 2017 - William Qiu

Conditional Processing
 Execute statements conditionally using IF-THEN logic.
 Control the length of character variables explicitly with
the LENGTH statement.
 Select rows to include in a SAS data set.

PSTAT 130 - Summer 2017 - William Qiu

Conditional Execution
General form of IF-THEN and ELSE statements:
IF expression THEN statement;
ELSE statement;
Expression contains operands and operators that form a set
of instructions that produce a value.
Examples:
If Hours < 40 then Status = 'Parttime';
If Jobcode = 'PILOT' then Bonus = Salary * 0.1;
else Bonus = 0;

Only one executable statement is allowed per IF-THEN or

ELSE statement.
PSTAT 130 - Summer 2017 - William Qiu
Conditional Execution
To allow more than one statement, use DO and END
statements.
IF expression THEN DO;
executable statements
END;
ELSE DO;
executable statements
END;

PSTAT 130 - Summer 2017 - William Qiu

 Subset Dataset
In a DATA step, you can subset the rows (observations) in a
SAS data set with a:
 WHERE statement
 DELETE statement
 subsetting IF statement.

The WHERE statement in a DATA step is the same as the

WHERE statement you saw in a PROC step.

PSTAT 130 - Summer 2017 - William Qiu

Selecting and Deleting Rows
Use an IF statement to include only those rows that meet
the criteria

IF expression;

Use an IF-THEN DELETE statement to exclude rows that

meet the criteria

IF expression THEN DELETE;

PSTAT 130 - Summer 2017 - William Qiu

WHERE versus IF
If the variable already exists in the input data set, you can
use a WHERE statement.
data work.tempdata;
set ia.empdata;
where jobcode = 'Pilot';
run;

If you are evaluating a calculated variable, use IF.

data work.tempdata;
set ia.empdata;
Bonus = Salary * 0.1;
if Bonus > $5000;
run;

PSTAT 130 - Summer 2017 - William Qiu

WHERE or Subsetting IF?
Step and Usage WHERE IF
PROC Step Yes No
DATA Step - Source of Variable
INPUT statement No Yes
(i.e., existing variable)
assignment statement No Yes
(i.e., created variable)
SET/MERGE (multiple data sets)
Variable in ALL data sets Yes Yes
Variable not in ALL data sets No Yes

PSTAT 130 - Summer 2017 - William Qiu

William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017 - William Qiu 2
Overview of 2nd Half of Quarter
 Combining Data Sets
 SET and MERGE
 Summarizing Data Sets
 PROC MEANS, FREQ, TABULATE, REPORT
 Graphing Data
 PROC GCHART, GPLOT
 Controlling Data Step Processing
 Do Loops – A Powerful Programming Tool
 Arrays and Macros

PSTAT 130 - Summer 2017 - William Qiu 3

Looking Behind the Scenes
The DATA step is processed in two phases:
 compilation
 execution
data work.dfwlax;
infile 'raw-data-file';
input Flight $ 1-3 Date $ 4-11
Dest $ 12-14 FirstClass 15-17
Economy 18-20;
run;

PSTAT 130 - Summer 2017 - William Qiu 4

Looking Behind the Scenes
At compile time, SAS creates
 an input buffer to hold the current raw data file record
that is being processed
1 2 3
1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0
0 1 5 1 0 / 2 5 / 1 2 L AX 14 1 6 3

 a program data vector (PDV) to hold the current SAS

observation
Flight $3 Date $8 Dest $3 FirstClass N8 Economy N8
015 10/25/12 LAX 14 163

 the descriptor portion of the output data set.

PSTAT 130 - Summer 2017 - William Qiu 5
Compiling the Data Step
SAS creates the data set
placeholder SAS opens the data source

data work.dfwlax;
infile 'raw-data-file';
input Flight $ 1-3 Date $ 4-11
Dest $ 12-14 FirstClass 15-17
Economy 18-20;
run;

SAS prepares the input buffer

PSTAT 130 - Summer 2017 - William Qiu 6

Lecture Overview
 A Look Behind Data Step Processing
 Compilation and Execution
 More on INPUT statements
 Examining Errors
 Data and Programming Errors

PSTAT 130 - Summer 2017 - William Qiu 7

DATA Step Execution: Summary

PSTAT 130 - Summer 2017 - William Qiu 8

Data Step Execution: Details
 At compilation, SAS creates an empty Input Buffer, and an
empty Program Data Vector, to store the incoming data
1 2 3 Empty Input
1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 Buffer

Flight $3 Date $8 Dest $3 FirstClass N 8 Economy N8 Empty Program

. .
Data Vector

 It then loads the first line of data into the input buffer,
parses it into variables, and outputs those values to the
SAS dataset Input Buffer with
1 2 3
1st line of data
1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0
0 1 5 1 0 / 2 5 / 1 2 L A X 1 4 1 6 3

Flight $3 Date $8 Dest $3 FirstClass N 8 Economy N8 Record output to

509 3/21/10 LAX 14 163
SAS Dataset

PSTAT 130 - Summer 2017 - William Qiu 9

Reading Data Using Formatted Input
Pointer controls:
 @n moves the pointer to column n.
 +n moves the pointer n positions.
An informat specifies
 the width of the input field
 how to read the data values that are stored in the field.

PSTAT 130 - Summer 2017 - William Qiu 10

Pointer Control - Absolute
 With formatted input, you can “point” at the first column
of each variable, instead of using start and end columns.
 The @ symbol tells SAS at what column to start reading the value

data students;
input @1 Name $8. @9 Gender $6. @18 Age 2. @22
Enroll mmddyy8.;
datalines;
David Male 19 06/18/10
Amelia Female 23 08/02/10
Ashley Female 20 09/14/10
Jim Male 26 08/26/10
;
run;

PSTAT 130 - Summer 2017 - William Qiu 11

Pointer Control - Relative
 You can also move the pointer forward a specific number
of spaces forward, using the + symbol
data students;
input Name $6. +2 Gender $6. +3 Age 2. +2 Enroll
mmddyy8.;
datalines;
David Male 19 06/18/10
Amelia Female 23 08/02/10
Ashley Female 20 09/14/10
Jim Male 26 08/26/10
;
run;

PSTAT 130 - Summer 2017 - William Qiu 12

Class Exercise
 Write an input statement to read in the following data,
which consists of five variables: instructor name,
academic rank, annual salary, course name, and first class
date.
John Tukey Asst $56,000 PSTAT130 09/23/10
Sigmund Freud Prof $92,000 PSYCH118 09/24/10
Karl Marx Assoc $78,000 POLI125 09/27/10

PSTAT 130 - Summer 2017 - William Qiu 13

Steps for Creating an Input Statement
1. How many variables are there in your data?
2. How many character variables? How many numeric
variables?
3. Are the data values separated by a single space (List input)
or are they presented in fixed columns (Column input)?
4. Are there any non-standard variables such as dates, commas
or dollar signs (Formatted input)?
5. Write an input statement with a name for each variable, and
a trailing dollar sign to indicate character variables.
6. For column input, add start and stop columns, or
absolute/relative pointers for each variable.
7. For formatted input, add appropriate character, numeric or
other formats to each variable.

PSTAT 130 - Summer 2017 - William Qiu 14

What Are Data Errors?
SAS detects data errors when
 the INPUT statement encounters invalid data in a field
 illegal arguments are used in functions
 impossible mathematical operations are requested.

PSTAT 130 - Summer 2017 - William Qiu 15

Examining Data Errors
When SAS encounters a data error,
1. a note that describes the error is printed in the SAS log
2. the input record being read is displayed in the SAS log
(contents of the input buffer)
3. the values in the SAS observation being created are
displayed in the SAS log (contents of the PDV)
4. a missing value is assigned to the appropriate SAS
variable
5. execution continues.

PSTAT 130 - Summer 2017 - William Qiu 16

Programming Errors - Tips
 Use the Enhanced Editor – it color-codes keywords and
highlights errors in red.
 Write your program in small parts and test each part.
 Clear the log and output before running your program
 Review the log, looking for red and green text.
 Confirm the number of records and variables in each data
set using the Log.
 Keep all variables in your interim data sets.
 Inspect the data sets you create in Table Editor, or using
Proc Print.

PSTAT 130 - Summer 2017 - William Qiu 17

William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017 - William Qiu
Lecture Outline
 Combining Datasets
 Concatenating Datasets (Appending)
 The SET statement
 Merging Datasets
 The MERGE and BY statements
 Types of Merging
 Parent-Child Relationships

PSTAT 130 - Summer 2017 - William Qiu

COMBINING SAS DATA SETS
 Append (or concatenate) – SET statement
 Two data sets each with the SAME variables
 Data set A has m records and k variables
 Data set B has n records and k variables
 Combined data set has (m + n) records and k variables
 Merge (or link) – use MERGE & BY statements
 Two data sets with at least one common variable and other
unique variables
 Data set A has m records and k UNIQUE variables
 Data set B has n records and j UNIQUE variables
 Combined data set (typically) has max(m,n) records and k+j+1
variables

PSTAT 130 - Summer 2017 - William Qiu

Concatenating SAS Data Sets
General form of a DATA step concatenation:

DATA SAS-data-set ;
SET SAS-data-set1 SAS-data-set2 . . .;
<other SAS statements>
RUN;

PSTAT 130 - Summer 2017 - William Qiu

Concatenating SAS Data Sets

SAS data sets work.qtr1

Jan data work.qtr1; Jan

set work.jan work.feb
Feb Feb
work.mar;
Mar
Mar run;

PSTAT 130 - Summer 2017 - William Qiu

Concatenating Data Sets – Example
I have two sections of my class, Morning and Afternoon,
and want to combine their scores
Data allsections;
Set morning afternoon;
run;

Morning Afternoon
Name Score allsections Name Score
Mary 75 Name Score Andy 78
Mark 82 Mary 75 Alice 85
Mike 68 Mark 82 Art 62
Mike 68
Andy 78
Alice 85
Art 62

PSTAT 130 - Summer 2017 - William Qiu

Concatenating Data Sets - Details
To append (or concatenate) data sets:
 The variable names and data types should be the same in
both data sets.
 You can append multiple data sets in a single SET
statement.
 You may want to create a variable that identifies the
source of each set of data. Do this in a separate data step
prior to appending the data sets.

PSTAT 130 - Summer 2017 - William Qiu

Creating an identifier variable
data morning;
input name $ score; Name Score Class
class = 'M'; Alice 85 A
datalines; … Art 62 A
run; Barry 82 M
Mary 75 M
data afternoon; Will 78 A
input name $ score; Zach 68 M
class = 'A';
datalines; …
run;

data allsections;
set morning afternoon;
run;

PSTAT 130 - Summer 2017 - William Qiu

Merging Data Sets
General form of a DATA step match-merge:

DATA SAS-data-set;
MERGE SAS-data-sets;
BY BY-variable(s);
<other SAS statements>
RUN;

Note: Data sets must be sorted on BY variable prior to

merging.

PSTAT 130 - Summer 2017 - William Qiu

Merging Data Sets - Example
I have midterm scores in one data set, and final scores in
another and want to combine them
Data allscores;
Merge midterm final;
by Name;
run;

Midterm Final
Name Midscore allsections Name Finalscore
Wendy 32 Name Midscore Finalscore John 91
Andy 38 Andy 38 82 Wendy 73
John 27 John 27 91 Andy 82
Wendy 32 73

PSTAT 130 - Summer 2017 - William Qiu

Way Have Separate Datasets?
 Efficiency of Storage
 Keep only the data you need in each data set
 Example: Customer information is kept separate from order
information
 Efficiency of Processing
 Smaller datasets can be processed faster than larger datasets
 Example: Reading in fewer variables, or sorting fewer records
 Different Data Sources
 Appending and Merging allow you to combine information from
separate sources
 Example: Information on the same topic from two different
researchers or two different websites

PSTAT 130 - Summer 2017 - William Qiu

The RENAME= Data Set Option
General form of the RENAME= data set option:
SAS-data-set(RENAME=(old-name-1=new-name-1
old-name-2=new-name-2
.
.
.
old-name-n=new-name-n))
 When appending data sets, use the RENAME= option to
create common variable names.
 When merging data sets, use the RENAME= option to
create unique variable names.
PSTAT 130 - Summer 2017 - William Qiu
RENAME: Appending Datasets
 Goal: append two datasets
 Problem: the variable names are different in the two
datasets (e.g., test scores are called “score” in Morning
section, and called “testscore” in Afternoon section.
 Solution: RENAME the testscore variable to score in the
Afternoon section
data allsections;
set morning
afternoon (RENAME=(testscore=score));
run;

PSTAT 130 - Summer 2017 - William Qiu

RENAME: Merging Datasets
 Goal: merge two datasets
 Problem: the variable names are the same in the two
datasets (e.g., midterm is called “score” in the Midterm
dataset, and Final is called “score” in Final data.
 Solution: RENAME score variable

Data allscores;
Merge midterm (RENAME=(score=midterm))
final (RENAME=(score=final));
by Name;
run;

PSTAT 130 - Summer 2017 - William Qiu

Types of Merges
 Match Merging – MERGE and BY statements
 The records from each data set with the same value of the
(unique) BY variable are linked, and output as one record
 If you omit the BY statement, the first record from each
data set are output together as one record without being
linked by a common variable
 One-to-many
 unique BY values are in one data set and duplicate
matching BY values are in the other data set.
 Many-to-many
 duplicate matching BY values are in both data sets.

PSTAT 130 - Summer 2017 - William Qiu

One-to-Many Merging

PSTAT 130 - Summer 2017 - William Qiu

One-to-Many Merging

PSTAT 130 - Summer 2017 - William Qiu

Many-to-Many Merging

PSTAT 130 - Summer 2017 - William Qiu

The IN= Data Set Option
Use the IN= option to create variables identifying which
data sets contained the observation.
General form of the IN= data set option:
SAS-data-set(IN=variable)

Example:
Data allscores;
Merge midterm (IN=InMidterm)
final (IN=InFinal);
by Name;
if InMidterm and InFinal;
run;
PSTAT 130 - Summer 2017 - William Qiu
Lookup Tables
 Data set variable contains ‘codes’
 Males are coded as 1
 Females are coded as 2
 Lookup table contains ‘labels’ that can be merged with
‘codes’

GenderLookup
GenderCode GenderLabel
1 Male
2 Female

PSTAT 130 - Summer 2017 - William Qiu

Parent – Child Tables
 Sales Order Form
 Parent Table contains Order-level information
 Order #, Order Date, Name, Phone Number, Shipping Address,
Billing Address, Payment Method, etc
 Child Table contains list of items ordered, quantity, unit cost
 Order # appears on every record in Parent and Child tables
 Tables are linked through One to Many merging using a
key variable such as Order #

PSTAT 130 - Summer 2017 - William Qiu

Scheduling Example
 The University maintains multiple data sets to schedule
classes
 A list of instructors and the courses they teach.
 A list of students taking each course.
 A list of classrooms and the courses that meet in them.

PSTAT 130 - Summer 2017 - William Qiu

Students 1 2
12345678901234567890
 Variables: John Thomas
Elizabeth Smith
 StudentName Rajesh Krish
 Three data sets are provided Lily Yang
Robert Williams
 PSTAT130.txt Tracy Jones
 PSYCH118.txt Cheryl Smith
Alex Shepard
 POLI125.txt Trinh Phan
Lee Barrett
Clark Johnson
Jenny Page
Mary Marcus
Curt Forrest
Andy Potts

PSTAT 130 - Summer 2017 - William Qiu

Instructors
 Variables:
 InstructorName
 AcademicRank
 Salary
 CourseName
 FirstClassDate
 File saved as Instructors.txt
1 2 3 4 5
12345678901234567890123456789012345678901234567890
John Tukey Assoc $56,000 PSTAT130 09/23/10
Sigmund Freud Assoc $92,000 PSYCH118 09/24/10
Karl Marx Asst $78,000 POLI125 09/27/10

PSTAT 130 - Summer 2017 - William Qiu

Classrooms
 Variables
 BldgName
 RoomNumber
 CourseName
 Days
 Time (read as a character variable)
 File saved as Classrooms.txt
1 2 3 4
123456789012345678901234567890123456789012345
Phelps Hall 222 PSYCH118 T/TH 10:00 am
South Hall 518 PSTAT130 M/W/F 5:00 pm
Phelps Hall 126 POLI125 M/W 2:00 pm

PSTAT 130 - Summer 2017 - William Qiu

Task - Input/Data Preparation
 Create three data steps to read in each list of students
 Create a variable to store the CourseName for each dataset
because only the file name indicates which class it is for
 Create a data step to read in Instructors data
 Create a data step to read in the Classrooms data
 Combine the student lists into a single data set of all
students - call it AllStudents
 Combine the Instructor, Classrooms and AllStudent data
sets so that each student is paired with his or her
instructor and the room information for that class

PSTAT 130 - Summer 2017 - William Qiu

Task – Create Roster
 Create a ‘Roster’ for each class showing the names of the
students taking that class.
 Separate the lists onto a page for each class.
 Assign appropriate variable labels for CourseName and
StudentName
 Show only the CourseName at the top of each page, and
the list of students names below.
------------------- Course Name=POLI125 ------------------
Student Name

Alex Shepard
Andy Potts
Cheryl Smith
Curt Forrest
PSTAT 130 - Summer 2017 - William Qiu
Task – Output Class List
 Create a Class List for each Student showing details of the
classes each is taking.
 Include the variables below
 Assign appropriate variable labels
 Use an appropriate format for FirstClassDate
 Example on next slide

PSTAT 130 - Summer 2017 - William Qiu

 Class List Example
------------------------------ Student Name=Alex Shepard --------------------------------

First
Course Class Instructor Building Room Class Class
Name Date Name Name Number Days Time

PSTAT130 09/23/10 John Tukey South Hall 518 M/W/F 5:00 pm

POLI125 09/27/10 Karl Marx Phelps Hall 126 M/W 2:00 pm

------------------------------- Student Name=Andy Potts ---------------------------------

First
Course Class Instructor Building Room Class Class
Name Date Name Name Number Days Time

PSTAT130 09/23/10 John Tukey South Hall 518 M/W/F 5:00 pm

PSYCH118 09/24/10 Sigmund Freud Phelps Hall 222 T/TH 10:00 am
POLI125 09/27/10 Karl Marx Phelps Hall 126 M/W 2:00 pm

PSTAT 130 - Summer 2017 - William Qiu

Tasks – Output
 Create a Master List of all Students whose instructor is an
Associate Professor.
 Include the variables in the example on the next page
 Assign appropriate variable labels
 Create a user-defined format and apply it to academic
rank, assigning the label “Assistant Professor” to “Asst”
and “Associate Professor” to “Assoc”
 Use an appropriate format for Salary

PSTAT 130 - Summer 2017 - William Qiu

Associate Professor Example
List of Students for Associate Professors

Course Instructor
Student Name Name Name Academic Rank Salary

Alex Shepard PSTAT130 John Tukey Associate Professor $56,000

Andy Potts PSTAT130 John Tukey Associate Professor $56,000
Andy Potts PSYCH118 Sigmund Freud Associate Professor $92,000
Cheryl Smith PSTAT130 John Tukey Associate Professor $56,000
Clark Johnson PSTAT130 John Tukey Associate Professor $56,000
Clark Johnson PSYCH118 Sigmund Freud Associate Professor $92,000
Curt Forrest PSTAT130 John Tukey Associate Professor $56,000
Curt Forrest PSYCH118 Sigmund Freud Associate Professor $92,000
Elizabeth Smith PSTAT130 John Tukey Associate Professor $56,000
Elizabeth Smith PSYCH118 Sigmund Freud Associate Professor $92,000
Jenny Page PSTAT130 John Tukey Associate Professor $56,000
Jenny Page PSYCH118 Sigmund Freud Associate Professor $92,000
John Thomas PSTAT130 John Tukey Associate Professor $56,000
John Thomas PSYCH118 Sigmund Freud Associate Professor $92,000
Lee Barrett PSTAT130 John Tukey Associate Professor $56,000
Lee Barrett PSYCH118 Sigmund Freud Associate Professor $92,000

PSTAT 130 - Summer 2017 - William Qiu

William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017 - William Qiu
Lecture Outline
 Summarizing Your Data
 PROC MEANS
 PROC FREQ
 PROC TABULATE
 PROC REPORT

PSTAT 130 - Summer 2017 - William Qiu

Summarizing Data
In order to summarize a data set, what three things do you
need to know?

PSTAT 130 - Summer 2017 - William Qiu

Summarizing Data
In order to summarize a data set, what three things do you
need to know?
 The Shape (Frequency Distribution, Outliers)
 The Center (Mean, Median, Mode)
 The Spread (Standard Deviation)

PSTAT 130 - Summer 2017 - William Qiu

Procedures for Summarizing Data
 PROC MEANS – calculate and display simple summary
statistics
 PROC FREQ – calculate and display frequency counts
 PROC TABULATE – flexible tool to calculate and display
multi-dimensional tables with summary statistics
 PROC REPORT – flexible tool for creating listings and
summary displays

PSTAT 130 - Summer 2017 - William Qiu

PROC MEANS Output

PSTAT 130 - Summer 2017 - William Qiu

PROC FREQ Output

PSTAT 130 - Summer 2017 - William Qiu

PROC TABULATE Output

PSTAT 130 - Summer 2017 - William Qiu

PROC REPORT Output

PSTAT 130 - Summer 2017 - William Qiu

Calculating Summary Statistics
PROC MEANS:
 calculates common summary statistics
 summarizes numeric variables
 BY and CLASS statements can be used to create
summaries for sub-groups
 can create an output data set of summary statistics

PSTAT 130 - Summer 2017 - William Qiu

Calculating Summary Statistics
 General form of a simple PROC MEANS step:

PROC MEANS DATA=SAS-data-set;

RUN;

 Example:

proc means data=ia.admit;

run;

PSTAT 130 - Summer 2017 - William Qiu

PROC MEANS – Default Output
The MEANS Procedure

Variable N Mean Std Dev Minimum Maximum

ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
Age 21 38.0476190 10.3124982 22.0000000 60.0000000
Date 21 14.5238095 9.1630729 1.0000000 31.0000000
Height 21 68.2380952 4.3116674 61.0000000 76.0000000
Weight 21 156.5238095 22.6398301 118.0000000 193.0000000
Fee 21 127.9500000 24.1524222 85.2000000 149.7500000
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

PSTAT 130 - Summer 2017 - William Qiu

Calculating Summary Statistics
By default, PROC MEANS
 analyzes every numeric variable in the SAS data set
 prints the statistics N, MEAN, STD, MIN, and MAX
 excludes missing values before calculating statistics.

PSTAT 130 - Summer 2017 - William Qiu

PROC MEANS – Options
 VAR <variable list> - selects variables to be
summarized
 BY <variable list> - creates separate summaries for
each BY group
 CLASS <variable list> - creates separate summaries
for each CLASS group
 OUTPUT out=sas data set – creates an output data
set containing summary statistics
 Statistical Keywords – option on PROC MEANS
statement
PSTAT 130 - Summer 2017 - William Qiu
Specifying Which Variables to Include
Use the VAR statement to select specific variables

proc means data=ia.admit;

var age height weight;
run;

PSTAT 130 - Summer 2017 - William Qiu

PROC MEAN - Common Statistics
 MIN
 MAX
 RANGE
 MEAN
 MEDIAN
 STDDEV
 SUM
 N
 NMISS – the number of observations missing a value on each
variable
 Confidence intervals, percentiles, and probability functions
can also be requested – see MEANS procedure, statistic
keywords in SAS Help

PSTAT 130 - Summer 2017 - William Qiu

Specifying Summary Statistics
List Keywords for statistics as Options to the PROC MEANS
statement

proc means data=ia.admit n mean stddev;

var age height weight;
run;

PSTAT 130 - Summer 2017 - William Qiu

Analyzing Subgroups – BY Statement
Use the BY statement to request summaries for sub-groups

proc means data=ia.admit n mean stddev;

var age height weight;
by actlevel;
output out=meansout;
run;

Data MUST be sorted on the BY variable(s) first

PSTAT 130 - Summer 2017 - William Qiu

BY Statement - Partial Output
The MEANS Procedure

-------------------- ActLevel=HIGH ---------------------

Variable N Mean Std Dev

------------------------------------------------------
Age 7 34.2857143 7.5213980
Height 7 70.1428571 4.2201332
Weight 7 163.5714286 21.1412483
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

--------------------- ActLevel=LOW ---------------------

Variable N Mean Std Dev

ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
Age 7 39.2857143 14.0441481
Height 7 66.4285714 5.0284903
Weight 7 150.7142857 26.1897835

PSTAT 130 - Summer 2017 - William Qiu

Analyzing Subgroups – CLASS Statement
Use the CLASS statement to request summaries for sub-
groups

proc means data=ia.admit n mean stddev;

class actlevel;
output out=meansout;
run;

Data DO NOT need to be sorted on the CLASS variable(s) first

PSTAT 130 - Summer 2017 - William Qiu

CLASS Statement - Ouput
PROC MEANS
With a CLASS Variable

The MEANS Procedure

Act N
Level Obs Variable N Mean Std Dev
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
HIGH 7 Age 7 34.2857143 7.5213980
Height 7 70.1428571 4.2201332
Weight 7 163.5714286 21.1412483

LOW 7 Age 7 39.2857143 14.0441481

Height 7 66.4285714 5.0284903
Weight 7 150.7142857 26.1897835

MOD 7 Age 7 40.5714286 8.6575043

Height 7 68.1428571 3.2877840
Weight 7 155.2857143 21.8305160
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

PSTAT 130 - Summer 2017 - William Qiu

Limit Number of Decimals
 Use the MAXDEC option in the PROC MEANS Statement
to limit the number of decimal places in the summary
statistics
proc means data=ia.admit n mean stddev maxdec=2;
var age height weight;
run;
Displays 2
The MEANS Procedure decimal places

Variable N Mean Std Dev

ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
Age 21 38.05 10.31
Height 21 68.24 4.31
Weight 21 156.52 22.64
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

PSTAT 130 - Summer 2017 - William Qiu

Class Exercise
 Open the heart dataset
 Inspect the dataset using PROC CONTENTS and PROC
PRINT
 Run PROC MEANS using default options
 Now limit the statistics to one decimal place
 Run PROC MEANS for the Arterial and Cardiac variables
 Run PROC MEANS using Sex as a Class variable

PSTAT 130 - Summer 2017 - William Qiu

Creating a Frequency Report
The FREQ procedure can be used to summarize the
frequency of values within the data set. In other words, it
creates frequency distributions.

PSTAT 130 - Summer 2017 - William Qiu

Creating a Frequency Report
 General form of a simple PROC FREQ step:

PROC FREQ DATA=SAS-data-set;

RUN;

 Example:

proc freq data=ia.admit;

run;

PSTAT 130 - Summer 2017 - William Qiu

Creating a Frequency Report
By default, PROC FREQ
 analyzes every variable in the SAS data set
 displays each distinct data value
 calculates the number of observations in which each data
value appears (and the corresponding relative and
cumulative percentages)
 indicates for each variable how many observations have
missing values.

PSTAT 130 - Summer 2017 - William Qiu

PROC FREQ – TABLES Statement
Use the TABLES statement to select variable and to specify
the type of frequency report.

General form of a PROC FREQ step with a TABLES

statement:
PROC FREQ DATA=SAS-data-set;
TABLES variable list;
RUN;

PSTAT 130 - Summer 2017 - William Qiu

 Creating a Frequency Report

Displays a frequency table

for the variable, Jobcode

The table lists each value of JobCode,

and its frequency

PSTAT 130 - Summer 2017 - William Qiu

Analyzing Categories of Values
Use the FORMAT Statement to analyze the frequency of
observations within User-Defined Categories

proc format;
value $codefmt
'FLTAT1'-'FLTAT3'='Flight Attendant'
'PILOT1'-'PILOT3'='Pilot';
run;
proc freq data = ia.crew;
format JobCode $codefmt.;
tables JobCode;
run;

PSTAT 130 - Summer 2017 - William Qiu

Analyzing Categories of Values

PSTAT 130 - Summer 2017 - William Qiu

Crosstabular Frequency Reports
General Form of two-way tables, known as crosstabs:

PROC FREQ DATA=SAS-data-set;

TABLES variable1 * variable 2;
RUN;

 Two-way tables categorize observations on the

combination of two sets of categories (e.g., Male Pilots,
Female Pilots, Male Flight Attendants and Female Flight
Attendantes)

PSTAT 130 - Summer 2017 - William Qiu

proc format;
value $codefmt
'FLTAT1'-'FLTAT3'='Flight Attendant'
'PILOT1'-'PILOT3'='Pilot';
value money
low-<25000 ='Less than 25,000'
25000-50000='25,000 to 50,000'
50000<-high='More than 50,000';
run;
proc freq data=ia.crew;
tables JobCode*Salary;
format JobCode $codefmt. Salary money.;
title 'Salary Distribution by Job Codes';
run;

PSTAT 130 - Summer 2017 - William Qiu

Crosstabular Output
Salary Distribution by Job Codes

The FREQ Procedure

Table of JobCode by Salary

JobCode Salary

Frequency ‚
Percent ‚
Row Pct ‚
Col Pct ‚Less tha‚25,000 t‚More tha‚ Total
‚n 25,000‚o 50,000‚n 50,000‚
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ
Flight Attendant ‚ 5 ‚ 39 ‚ 0 ‚ 44
‚ 7.25 ‚ 56.52 ‚ 0.00 ‚ 63.77
‚ 11.36 ‚ 88.64 ‚ 0.00 ‚
‚ 100.00 ‚ 100.00 ‚ 0.00 ‚
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ
Pilot ‚ 0 ‚ 0 ‚ 25 ‚ 25
‚ 0.00 ‚ 0.00 ‚ 36.23 ‚ 36.23
‚ 0.00 ‚ 0.00 ‚ 100.00 ‚
‚ 0.00 ‚ 0.00 ‚ 100.00 ‚
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ
Total 5 39 25 69
7.25 56.52 36.23 100.00
PSTAT 130 - Summer 2017 - William Qiu
William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017 - William Qiu
Lecture Outline
 Summarizing Your Data – con’t
 PROC MEANS
 PROC FREQ
 PROC TABULATE
 PROC REPORT

PSTAT 130 - Summer 2017 - William Qiu

Procedures for Summarizing Data
 PROC MEANS – calculate and display simple summary
statistics
 PROC FREQ – calculate and display frequency counts
 PROC TABULATE – flexible tool to calculate and display
multi-dimensional tables with summary statistics
 PROC REPORT – flexible tool for creating listings and
summary displays

PSTAT 130 - Summer 2017 - William Qiu

PROC MEANS Output

PSTAT 130 - Summer 2017 - William Qiu

PROC FREQ Output

PSTAT 130 - Summer 2017 - William Qiu

PROC TABULATE Output

PSTAT 130 - Summer 2017 - William Qiu

PROC REPORT Output

PSTAT 130 - Summer 2017 - William Qiu

The TABULATE Procedure
The report writing features of PROC TABULATE include
 control of table construction
 differentiating between classification variables and
analysis variables
 specifying statistics
 formatting of values
 labeling variables and statistics.

PSTAT 130 - Summer 2017 - William Qiu

PROC TABULATE syntax
General form of a PROC TABULATE step

PROC TABULATE DATA=SAS-data-set <options>;

CLASS class-variables;
VAR analysis-variables;
TABLE page-expression,
row-expression,
column-expression </ option(s)>;
RUN;

PSTAT 130 - Summer 2017 - William Qiu

Specifying Classification Variables

PROC TABULATE DATA=SAS-data-set <options>;

CLASS class-variables;
VAR analysis-variables;
TABLE page-expression,
row-expression,
column-expression </ option(s)>;
RUN;

 Class variables are used to define subgroups on one or

more dimensions

PSTAT 130 - Summer 2017 - William Qiu

Specifying Analysis Variables

PROC TABULATE DATA=SAS-data-set <options>;

CLASS class-variables;
VAR analysis-variables;
TABLE page-expression,
row-expression,
column-expression </ option(s)>;
RUN;

 Summary statistics (e.g., means) are calculated for the

VAR variables

PSTAT 130 - Summer 2017 - William Qiu

Specifying Table Structure

PROC TABULATE DATA=SAS-data-set <options>;

CLASS class-variables;
VAR analysis-variables;
TABLE page-expression,
row-expression,
column-expression </ option(s)>;
RUN;

 The TABLE statement specifies the format of the table

PSTAT 130 - Summer 2017 - William Qiu

Using Only Class Variables
title 'Flight Attendant Counts by Location';
proc tabulate data=ia.fltat;
class Location;
table Location;
run;

Flight Attendant Counts by Location

„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ†
‚ Location ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ CARY ‚ FRANKFURT ‚ LONDON ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ N ‚ N ‚ N ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ 17.00‚ 12.00‚ 15.00‚
Šƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ
PSTAT 130 - Summer 2017 - William Qiu
Obtaining a Total
proc tabulate data=ia.fltat;
class Location;
table Location All
run;
Blank Operator between Location and
All concatenates information

Flight Attendant Counts by Location

„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ†
‚ Location ‚ ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚
‚ CARY ‚ FRANKFURT ‚ LONDON ‚ All ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ N ‚ N ‚ N ‚ N ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ 17.00‚ 12.00‚ 15.00‚ 44.00‚
Šƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ
PSTAT 130 - Summer 2017 - William Qiu
Two-Dimensional Tables
title2 'by JobCode';
proc tabulate data=ia.fltat;
class Location JobCode;
table JobCode, Location;
run;

Column Dimension
Comma operator
Row Dimension moves to a new
dimension

PSTAT 130 - Summer 2017 - William Qiu

Two-Dimensional Tables
Flight Attendant Counts by Location
by JobCode
Column Dimension
„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ†
‚
Row Dimension‚ Location ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ CARY ‚ FRANKFURT ‚ LONDON ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ N ‚ N ‚ N ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚JobCode ‚ ‚ ‚ ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚ ‚ ‚
‚FLTAT1 ‚ 5.00‚ 4.00‚ 5.00‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT2 ‚ 7.00‚ 5.00‚ 6.00‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT3 ‚ 5.00‚ 3.00‚ 4.00‚
Šƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ

PSTAT 130 - Summer 2017 - William Qiu

Subsetting the Data

title 'Counts for Cary and Frankfurt';

proc tabulate data=ia.fltat;
where Location in ('CARY', 'FRANKFURT');
class Location JobCode;
table JobCode, Location;
run;

PSTAT 130 - Summer 2017 - William Qiu

Subsetting the Data
Flight Attendant Counts by Location
by JobCode

„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ†
‚ ‚ Location ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ CARY ‚ FRANKFURT ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ N ‚ N ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚JobCode ‚ ‚ ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚ ‚
‚FLTAT1 ‚ 5.00‚ 4.00‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT2 ‚ 7.00‚ 5.00‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT3 ‚ 5.00‚ 3.00‚
Šƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ

PSTAT 130 - Summer 2017 - William Qiu

Two-Dimensional Tables

proc tabulate data=ia.fltat;

where Location in ('CARY', 'FRANKFURT');
class Location JobCode;
table JobCode all, Location all;
run;

Row Dimension
Column Dimension

PSTAT 130 - Summer 2017 - William Qiu

Two-Dimensional Table
Total Salary for Cary and Frankfurt All in Column
Dimension
„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ†
‚ ‚ Location ‚ ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚
All in Row
‚ ‚ CARY ‚ FRANKFURT ‚ All ‚
Dimension
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ N ‚ N ‚ N ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚JobCode ‚ ‚ ‚ ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚ ‚ ‚
‚FLTAT1 ‚ 5.00‚ 4.00‚ 9.00‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT2 ‚ 7.00‚ 5.00‚ 12.00‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT3 ‚ 5.00‚ 3.00‚ 8.00‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚All ‚ 17.00‚ 12.00‚ 29.00‚
Šƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ

PSTAT 130 - Summer 2017 - William Qiu

Using Analysis Variables

title 'Total Salary for Cary and Frankfurt';

proc tabulate data=ia.fltat;
where Location in ('CARY', 'FRANKFURT');
class Location JobCode;
var Salary;
table JobCode, Location*Salary;
run;

PSTAT 130 - Summer 2017 - William Qiu

Using Analysis Variables Salary within
Total Salary for Cary and Frankfurt each Location

„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ†
‚ ‚ Location ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ CARY ‚ FRANKFURT ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ Salary ‚ Salary ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ Sum ‚ Sum ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚JobCode ‚ ‚ ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚ ‚
‚FLTAT1 ‚ 131000.00‚ 100000.00‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT2 ‚ 245000.00‚ 181000.00‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT3 ‚ 217000.00‚ 134000.00‚
Šƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ

PSTAT 130 - Summer 2017 - William Qiu

Formatting the Statistic

proc tabulate data=ia.fltat format=dollar12.;

where Location in ('CARY', 'FRANKFURT');
class Location JobCode;
var Salary;
table JobCode, Location*Salary;
run;

PSTAT 130 - Summer 2017 - William Qiu

Formatting a Statistic
Total Salary for Cary and Frankfurt
FORMAT= Option
changes default
„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ† format for all cells
‚ ‚ Location ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ CARY ‚ FRANKFURT ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ Salary ‚ Salary ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ Sum ‚ Sum ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚JobCode ‚ ‚ ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚ ‚
‚FLTAT1 ‚ $131,000‚ $100,000‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT2 ‚ $245,000‚ $181,000‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT3 ‚ $217,000‚ $134,000‚
Šƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ

PSTAT 130 - Summer 2017 - William Qiu

Specifying a Statistic
title 'Average Salary for Cary and Frankfurt';
proc tabulate data=ia.fltat format=dollar12.;
where Location in ('CARY', 'FRANKFURT');
class Location JobCode;
var Salary;
table JobCode, Location*Salary*mean;
run;

PSTAT 130 - Summer 2017 - William Qiu

Specifying a Statistic
Average Salary for Cary and Frankfurt
MEAN
„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ† statistic
‚ ‚ Location ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ CARY ‚ FRANKFURT ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ Salary ‚ Salary ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ Mean ‚ Mean ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚JobCode ‚ ‚ ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚ ‚
‚FLTAT1 ‚ $26,200‚ $25,000‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT2 ‚ $35,000‚ $36,200‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT3 ‚ $43,400‚ $44,667‚
Šƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ

PSTAT 130 - Summer 2017 - William Qiu

ALL with Analysis Variable
General form for generating overall information when using
an analysis variable:
ALL*analysis-variable*statistic keyword

proc tabulate data=ia.fltat format=dollar12.;

where Location in ('CARY', 'FRANKFURT');
class Location JobCode;
var Salary;
table JobCode all,
Location*Salary*mean all*Salary*mean;
run;

PSTAT 130 - Summer 2017 - William Qiu

proc tabulate data=ia.fltat format=dollar12.;
where Location in ('CARY', 'FRANKFURT');
ALL with Analysis Variable
class Location JobCode;
var Salary;
table JobCode all,
Location*Salary*mean Averageall*Salary*mean;
Salary for Cary and Frankfurt
run;
„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ†
‚ ‚ Location ‚ ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚
‚ ‚ CARY ‚ FRANKFURT ‚ All ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ Salary ‚ Salary ‚ Salary ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚ ‚ Mean ‚ Mean ‚ Mean ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚JobCode ‚ ‚ ‚ ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚ ‚ ‚
‚FLTAT1 ‚ $26,200‚ $25,000‚ $25,667‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT2 ‚ $35,000‚ $36,200‚ $35,500‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT3 ‚ $43,400‚ $44,667‚ $43,875‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚All ‚ $34,882‚ $34,583‚ $34,759‚
Šƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ

PSTAT 130 - Summer 2017 - William Qiu

proc tabulate data=ia.fltat format=dollar12.;
where Location in ('CARY', 'FRANKFURT');
PROC TABULATE Review
class Location JobCode;
var Salary;
Column
Dimension
table JobCode all,
Location*Salary*mean Averageall*Salary*mean;
Salary for Cary and Frankfurt
run; All in Column
„ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ† Dimension
‚ ‚ Location ‚ ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒ…ƒƒƒƒƒƒƒƒƒƒƒƒ‰
Salary within ‚
‚ ‚ CARY ‚ FRANKFURT ‚ All ‚
Row
‚
each Location
‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
Dimension
‚ ‚ Salary ‚ Salary ‚ Salary ‚
‚ ‡ƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰ FORMAT=
‚ ‚ Mean ‚ Mean ‚ Mean ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
option
‚JobCode ‚ ‚ ‚ ‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‰ ‚ ‚ ‚
‚FLTAT1 ‚ $26,200‚ $25,000‚ $25,667‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT2 ‚ $35,000‚ $36,200‚ $35,500‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚FLTAT3 ‚ $43,400‚ $44,667‚ $43,875‚
‡ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒƒƒƒƒ‰
‚All ‚ $34,882‚ $34,583‚ $34,759‚
Šƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒ‹ƒƒƒƒƒƒƒƒƒƒƒƒŒ
All in Row
Dimension PSTAT 130 - Summer 2017 - William Qiu
The REPORT Procedure
PROC REPORT enables you to
 create listing reports
 create summary reports
 enhance reports
 request separate subtotals and grand totals
 generate reports in an interactive point-and-click or
programming environments.

PSTAT 130 - Summer 2017 - William Qiu

PROC REPORT versus PROC PRINT

PSTAT 130 - Summer 2017 - William Qiu

Creating a Listing Report
General form of a simple PROC REPORT step:
PROC REPORT DATA=SAS-data-set <options>;
RUN;

Example:
proc report data=ia.admit nowindow;
run;

Note: the NoWindow option suppresses the interactive report writing

function.

PSTAT 130 - Summer 2017 - William Qiu

PROC REPORT Default Output
The SAS System 10:23 Monday, November 7, 2011 3

Act
ID Name Sex Age Date Height Weight Level Fee
2458 Murray, W M 27 1 72 168 HIGH 85.20
2462 Almers, C F 34 3 66 152 HIGH 124.80
2501 Bonaventure, T F 31 17 61 123 LOW 149.75
2523 Johnson, R F 43 31 63 137 MOD 149.75
2539 LaMance, K M 51 4 71 158 LOW 124.80
2544 Jones, M M 29 6 76 193 HIGH 124.80
2552 Reberson, P F 32 9 67 151 MOD 149.75
2555 King, E M 35 13 70 173 MOD 149.75
2563 Pitts, D M 34 22 73 154 LOW 124.80
2568 Eberhardt, S F 49 27 64 172 LOW 124.80
2571 Nunnelly, A F 44 19 66 140 HIGH 149.75
2572 Oberon, M F 28 17 62 118 LOW 85.20
2574 Peterson, V M 30 6 69 147 MOD 149.75
2575 Quigley, M F 40 8 69 163 HIGH 124.80
2578 Cameron, L M 47 5 72 173 MOD 124.80
2579 Underwood, K M 60 22 71 191 LOW 149.75
2584 Takahashi, Y F 43 29 65 123 MOD 124.80
2586 Derber, B M 25 23 75 188 HIGH 85.20
2588 Ivan, H F 22 20 63 139 LOW 85.20
2589 Wilcox, E F 41 16 67 141 HIGH 149.75
2595 Warren, C M 54 7 71 183 MOD 149.75

PSTAT 130 - Summer 2017 - William Qiu

The REPORT Procedure
The default listing displays
 each data value as it is stored in the data set, or
formatted value if a format is stored with the data
 variable names or labels as report column headings
 a default width for the report columns
 character values left-justified
 numeric values right-justified
 observations in the order in which they are stored in the
data set.

PSTAT 130 - Summer 2017 - William Qiu

Printing Selected Variables
General form of the COLUMN statement:

COLUMN SAS-variables;

PSTAT 130 - Summer 2017 - William Qiu

Sample Listing Report
title 'Salary Analysis';
proc report data=ia.crew nowd;
column JobCode Location Salary;
run;

Partial JobCode Location Salary

SAS PILOT1 LONDON 72000
FLTAT3 CARY 41000
Output PILOT2 FRANKFURT 81000
PILOT2 FRANKFURT 83000
FLTAT2 LONDON 36000
PILOT1 LONDON 65000
FLTAT2 FRANKFURT 35000
FLTAT2 FRANKFURT 38000
FLTAT1 LONDON 28000
FLTAT3 LONDON 44000
FLTAT2 CARY 37000

PSTAT 130 - Summer 2017 - William Qiu

The DEFINE Statement
You can enhance the report by using DEFINE statements to
 define how each variable is used in the report
 assign formats to variables
 specify report column headers and column widths
 change the order of the rows in the report.

PSTAT 130 - Summer 2017 - William Qiu

The DEFINE Statement - Details
 Character variables are used as Display variables
 Numeric variabeles are used as Analysis variables
 FORMAT= option assigns format to a variable
 WIDTH= option controls width of report column
 ‘report-column-header’ – defines the column header for
a variable

PSTAT 130 - Summer 2017 - William Qiu

Enhanced Listing Report
proc report data=ia.crew nowd;
column JobCode Location Salary;
define JobCode / width=8 'Job Code';
define Location / 'Home Base';
define Salary / format=dollar10.;
run;

Partial Job Code Home Base Salary

SAS PILOT1 LONDON $72,000
Output FLTAT3 CARY $41,000
PILOT2 FRANKFURT $81,000
PILOT2 FRANKFURT $83,000
FLTAT2 LONDON $36,000
PILOT1 LONDON $65,000
FLTAT2 FRANKFURT $35,000
FLTAT2 FRANKFURT $38,000
FLTAT1 LONDON $28,000
FLTAT3 LONDON $44,000
PSTAT 130 - Summer 2017 - William Qiu
ORDER Usage Type
 ORDER keyword identifies variable used to order rows of
report
proc report data=ia.crew nowd;
column JobCode Location Salary;
define JobCode / order width=8 'Job Code';
define Location / 'Home Base';
define Salary / format=dollar10.;
run;

PSTAT 130 - Summer 2017 - William Qiu

ORDER Usage Type
Job Code Home Base Salary
Partial FLTAT1 LONDON $28,000
SAS FRANKFURT $25,000
Output CARY $23,000
CARY $21,000
CARY $28,000
FRANKFURT $22,000
LONDON $29,000
FLTAT2 LONDON $36,000 Individual Values
FRANKFURT $35,000 (not Summary Stats)
FRANKFURT $38,000
CARY $37,000
CARY $34,000
LONDON $34,000
CARY $36,000
FLTAT3 CARY $41,000
LONDON $44,000
FRANKFURT $48,000
FRANKFURT $45,000
CARY $44,000
PSTAT 130 - Summer 2017 - William Qiu
Defining Group Variables
Use the REPORT procedure to create a summary report by
defining variables as group variables.

All observations whose group variables have the same

values are collapsed into a single row in the report.

PSTAT 130 - Summer 2017 - William Qiu

Defining Group Variables
If you have a group variable, there must be no display or
order variables.
 Group variables produce summary reports (observations
collapsed into groups).
 Display and order variables produce listing reports (one
row for each observation).

PSTAT 130 - Summer 2017 - William Qiu

Defining Analysis Variables
Default usage for numeric variables is ANALYSIS with a
default statistic of SUM.
 If the report contains group variables, the report displays
the sum of the numeric variables’ values for each group.
 If the report contains at least one display or order
variable and no group variables, the report lists all of the
values of the numeric variable.
 If the report contains only numeric variables, the report
displays grand totals for the numeric variables.

PSTAT 130 - Summer 2017 - William Qiu

Defining Analysis Variables
 Selected statistics include:
SUM sum (default)
N number of non-missing values
MEAN Average
MAX Maximum Value
MIN Minumum Value

 Example:

define Salary / mean format=dollar10.;

PSTAT 130 - Summer 2017 - William Qiu

Summarizing the Data
Use the GROUP usage in the DEFINE statement to specify
the variables that define groups.

proc report data=ia.crew nowd;

column JobCode Location Salary;
define JobCode / group width=8 'Job Code';
define Location / group 'Home Base';
define Salary / analysis mean format=dollar10.;
run;

PSTAT 130 - Summer 2017 - William Qiu

Summarizing the Data
Salary Analysis

Job Code Home Base Salary

FLTAT1 CARY $26,200
FRANKFURT $25,000
LONDON $25,600
FLTAT2 CARY $35,000
FRANKFURT $36,200
LONDON $34,333 Mean Salary for Flight
FLTAT3 CARY $43,400 Attendant 1’s at
FRANKFURT $44,667 London Base
LONDON $45,000
PILOT1 CARY $70,333
FRANKFURT $67,500
LONDON $70,000
PILOT2 CARY $80,750
FRANKFURT $80,000
LONDON $79,000
PILOT3 CARY $100,000
FRANKFURT $102,500
LONDON $98,000

PSTAT 130 - Summer 2017 - William Qiu

Printing Grand Totals
General form of the RBREAK statement:
RBREAK BEFORE|AFTER <options>;

Selected options:
SUMMARIZE Prints the total.
OL Prints a single line above the total.
DOL Prints a double line above the total.
UL Prints a single line below the total.
DUL Prints a double line below the total.

PSTAT 130 - Summer 2017 - William Qiu

The BREAK and RBREAK Statements
Use the BREAK statement to display a total at the end of a
Group. Use the RBREAK statement to display the grand
total at the bottom of the report.

proc report data=ia.crew nowd;

column JobCode Location Salary;
define JobCode / group width=8 'Job Code';
define Location / group 'Home Base';
define Salary / format=dollar10.;
rbreak after / summarize dol;
run;

PSTAT 130 - Summer 2017 - William Qiu

The RBREAK Statement
Job Code Home Base Salary
FLTAT1 CARY $131,000
FRANKFURT $100,000
LONDON $128,000
FLTAT2 CARY $245,000
FRANKFURT $181,000
LONDON $206,000
FLTAT3 CARY $217,000
FRANKFURT $134,000
LONDON $180,000
PILOT1 CARY $211,000
FRANKFURT $135,000
LONDON $210,000
PILOT2 CARY $323,000
FRANKFURT $240,000
LONDON $158,000
PILOT3 CARY $300,000 DOL - Double Overline
FRANKFURT $205,000
LONDON $294,000
RBREAK (Grand Total) ==========
$3,598,000

PSTAT 130 - Summer 2017 - William Qiu

Enhancing the Report
proc report data=ia.crew nowd headline headskip;
column JobCode Location Salary;
define JobCode / group width=8 'Job Code';
define Location / group 'Home Base';
define Salary / format=dollar10.;
rbreak after / summarize dol;
run;

PSTAT 130 - Summer 2017 - William Qiu

Enhancing the Report Head Line
Job Code Home Base Salary
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

FLTAT1 CARY $131,000 Head Skip

FRANKFURT $100,000 (Blank Line)
LONDON $128,000
FLTAT2 CARY $245,000
FRANKFURT $181,000
LONDON $206,000
FLTAT3 CARY $217,000
FRANKFURT $134,000
LONDON $180,000
PILOT1 CARY $211,000
FRANKFURT $135,000
LONDON $210,000
PILOT2 CARY $323,000
FRANKFURT $240,000
LONDON $158,000
PILOT3 CARY $300,000
FRANKFURT $205,000
LONDON $294,000
==========
$3,598,000

PSTAT 130 - Summer 2017 - William Qiu

William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017- William Qiu
PROC TABULATE vs PROC REPORT

PSTAT 130 - Summer 2017- William Qiu

Lecture Outline
 Producing HTML output using the Output Deliver System
 Summaring data with graphs
 Bar Charts – vertical and horizontal
 Pie Charts
 Graphing averages and sums
 Graphing options

PSTAT 130 - Summer 2017- William Qiu

HTML Output
New to SAS Version 9.3
 Output can be directed to
 the Output Window (in text format) or
 the Results Viewer (in HTML format) or both
 Select using the menu command:
Tools | Options | Preferences |Results

PSTAT 130 - Summer 2017- William Qiu

The Output Delivery System

PSTAT 130 - Summer 2017- William Qiu

Generating an HTML File
The ODS HTML statement opens, closes, and manages the
HTML destination.
General form of the ODS HTML statement:

ODS HTML FILE='HTML-file-specification' <options>;

SAS code that generates output
ODS HTML CLOSE;

PSTAT 130 - Summer 2017- William Qiu

Generating an HTML File
Output is directed to the specified HTML file until you
 close the HTML destination
 specify another destination file.
HTML File

ods html file='…'; Report

proc print…
proc means… Report
proc freq…
ods html close; Report

PSTAT 130 - Summer 2017- William Qiu

Creating an HTML Report
1. Open an HTML destination for the output.
2. Generate the output.
3. Close the HTML destination.

ods html file='Salary.html';

proc print data=ia.empdata label noobs;
label Salary='Annual Salary';
format Salary money. Jobcode $codefmt.;
title1 'Salary Report';
run;
ods html close;

PSTAT 130 - Summer 2017- William Qiu

Creating an HTML Report

PSTAT 130 - Summer 2017- William Qiu

Apply an ODS Style
ODS Styles are pre-defined formats for output.

 Example:
ods html file=‘output.html'
style=analysis;

 Complete List of Styles:

proc template;
list styles;
run;
quit;

PSTAT 130 - Summer 2017- William Qiu

ODS File Formats
With ODS you can create file formats:
 HTML – HyperText Markup Language – for web pages
 RTF – Rich Text Format – for Word
 PDF – Portable Document Format – for Adobe
 PS – Post-Script – for printers
 Excel
 XML
 and many others

PSTAT 130 - Summer 2017- William Qiu

Summarizing Data with Graphs
You can use bar or pie charts to graphically display the
following:
 distribution of a variable’s values
 average value of a variable for different categories
 total value of a variable for different categories

PSTAT 130 - Summer 2017- William Qiu

Vertical Bar Chart

PSTAT 130 - Summer 2017- William Qiu

Horizontal Bar Chart

PSTAT 130 - Summer 2017- William Qiu

Pie Chart

PSTAT 130 - Summer 2017- William Qiu

Specifying a Chart
Use the GCHART procedure to do the following:
 Specify the physical form of the chart.
 Identify a chart variable that determines the number of
bars or pie slices to create.
 Optionally identify an analysis variable to use for
calculating statistics that determine the height (or length)
of the bar or the size of the slice.

PSTAT 130 - Summer 2017- William Qiu

The GCHART Procedure
 General form of the PROC GCHART statement:

PROC GCHART DATA=SAS-data-set;

 Use one of these statements to specify the desired type

of chart:

HBAR chart-variable . . . </options>;

VBAR chart-variable . . . </options>;
PIE chart-variable . . . </options>;

PSTAT 130 - Summer 2017- William Qiu

Chart Variable
The chart variable
 determines the number of bars or slices produced within
a graph
 can be character or numeric.

PSTAT 130 - Summer 2017- William Qiu

Basic Vertical Bar Chart
Produce a vertical bar chart that displays the number of
employees in each job code.

proc gchart data=ia.crew;

vbar JobCode;
run;

PSTAT 130 - Summer 2017- William Qiu

Basic Vertical Bar Chart

PSTAT 130 - Summer 2017- William Qiu

Defining the Graphics Device
General form of the GOPTIONS statement:
GOPTIONS graphics-options;

Devices control the exact format of the graph and can be

either printers or application drivers

PSTAT 130 - Summer 2017- William Qiu

Vertical Bar Chart with ODS HTML
Modify the vertical bar chart to route the output to a Java
image within an HTML document.
ods html file='vbar.html';
goptions dev=activex;

proc gchart data=ia.crew;

vbar JobCode;
run;

ods html close;

goptions reset=all;

PSTAT 130 - Summer 2017- William Qiu

Vertical Bar Chart with ODS HTML

PSTAT 130 - Summer 2017- William Qiu

Horizontal Bar Chart
Produce a horizontal bar chart that displays the number of
employees in each job code.

proc gchart data=ia.crew;

hbar JobCode;
run;

PSTAT 130 - Summer 2017- William Qiu

Horizontal Bar Chart

PSTAT 130 - Summer 2017- William Qiu

Pie Chart
Produce a pie chart that displays the number of employees
in each job code. Route the output to an ActiveX image in
an HTML document.
ods html file='piechart.html';
goptions dev=activex;

proc gchart data=ia.crew;

pie JobCode;
run;

ods html close;

goptions reset=all;

PSTAT 130 - Summer 2017- William Qiu

Pie Chart

PSTAT 130 - Summer 2017- William Qiu

Charting Character Variables

PSTAT 130 - Summer 2017- William Qiu

Charting Numeric Variables
 GCHART can graph character or numeric variables
 When graphing numeric variables, by default, the values
are grouped together to create a smaller number of lines
 To override the default behavior for numeric chart
variables, use the DISCRETE option in the HBAR, VBAR, or
PIE statement.

PSTAT 130 - Summer 2017- William Qiu

Numeric Chart – Default Grouping

PSTAT 130 - Summer 2017- William Qiu

Numeric Chart – DISCRETE option

PSTAT 130 - Summer 2017- William Qiu

Charting Summary Statistics
 You can chart a summary statistic (e.g., mean) using the
SUMVAR= and TYPE= options
 General form:

 Description:
SUMVAR = identifies the analysis variable to use for
the sum or mean calculation.
TYPE= specifies that the height or length of the
bar or size of the slice represents a mean
or sum of the analysis-variable values.

PSTAT 130 - Summer 2017- William Qiu

Using an Analysis Variable
Produce a vertical bar chart that displays the average salary
of employees in each job code.
ods html file='vbar2.html';
goptions dev=actximg;

proc gchart data=ia.crew;

vbar JobCode / sumvar=Salary
type=mean;
run;

ods html close;

Goptions reset=all;

PSTAT 130 - Summer 2017- William Qiu

Using an Analysis Variable

PSTAT 130 - Summer 2017- William Qiu

FILL Option
goptions dev=gif;
proc gchart data=ia.crew;
pie JobCode / sumvar=Salary type=sum
fill=x;
format Salary dollar8.;
run;
goptions reset=all;

PSTAT 130 - Summer 2017- William Qiu

EXPLODE Option
goptions dev=gif;
proc gchart data=ia.crew;
pie JobCode / sumvar=Salary type=sum
explode = 'PILOT3';
format Salary dollar8.;
run;
goptions reset=all;

PSTAT 130 - Summer 2017- William Qiu

Adding Titles and Footnote
 Selected options

COLOR=color | C=color
FONT=type-font | F=type-font
HEIGHT=n | H=n

PSTAT 130 - Summer 2017- William Qiu

Title and Footnote Options

PSTAT 130 - Summer 2017- William Qiu

Producing Plots
Use the GPLOT procedure to produce scatterplots and line
graphs
 General form:
PROC GPLOT DATA=SAS-data-set;
PLOT vertical-variable*horizontal-variable </options>;
RUN;
QUIT;

 Example:

proc gplot data=data1.admit;

plot weight * height;
run;

PSTAT 130 - Summer 2017- William Qiu

Producing Line Graph
Produce a plot of the number of passengers by date for
flight number 114 over a one-week period.
ods html file='plot.html';
ods graphics;

proc gplot data=data1.flight114;

where date between '02mar2001'd and
'08mar2001'd; *this selects one week of flights;

plot Boarded*Date;

run;

goptions reset=all;
PSTAT 130 - Summer 2017- William Qiu
Producing Plots – GPLOT Output

PSTAT 130 - Summer 2017- William Qiu

SYMBOL Statement
You can use the SYMBOL statement to do the following:
 define plotting symbols
 draw lines through the data points
 specify the color of the plotting symbols and lines

 General Form
SYMBOLn options;

PSTAT 130 - Summer 2017- William Qiu

Symbol Statement Options
Use a square as the plotting symbol and join the points with
straight lines.
proc gplot data=ia.flight114;
where date between '02mar2001'd and
'08mar2001'd;
plot Boarded*Date;
symbol value=square i=join;
run;

PSTAT 130 - Summer 2017- William Qiu

SYMBOL Statement
SYMBOL statements have the following characteristics:
global After they are defined, they remain in
effect until changed or until the end of
the SAS session.
additive Specifying the value of one option does
not affect the values of other options.

PSTAT 130 - Summer 2017- William Qiu

Symbol Statement Options
 Options for shape of symbol
VALUE=symbol | V=symbol

 Selected symbol values include the following:

PSTAT 130 - Summer 2017- William Qiu

Symbol Statement Options
 Interpolation
I=interpolation

 Selected interpolation values:

 Combining symbol value=none with interpolation

produces a line-only plot

PSTAT 130 - Summer 2017- William Qiu

Additional SYMBOL Statement Options

proc gplot data=ia.flight114;

where date between '02mar2001'd and
'08mar2001'd;
plot Boarded*Date;
symbol color=red width=2;
run;

PSTAT 130 - Summer 2017- William Qiu

Additional SYMBOL Statement Options

PSTAT 130 - Summer 2017- William Qiu

Modifying the SYMBOL Statement
 Set the attributes only for SYMBOL1.
symbol1 c=blue v=diamond;
 Modify only the color of SYMBOL1, not the V= option
setting.
symbol1 c=green;

 To cancel SYMBOL Staments:

symbol1;
-OR-
goptions reset=symbol;

PSTAT 130 - Summer 2017- William Qiu

Controlling the Axis Appearance
You can modify the appearance of the axes that PROC
GPLOT produces with the following:
 PLOT statement options
 the LABEL statement
 the FORMAT statement

PSTAT 130 - Summer 2017- William Qiu

PLOT Statement Options
 Define the scale on the vertical axis and display the axis
text in blue.
proc gplot data=ia.flight114;
where date between '02mar2001'd and
'08mar2001'd;
plot Boarded*Date / vaxis=100 to 200 by 25
ctext=blue;
symbol value=square i=join;
run;

PSTAT 130 - Summer 2017- William Qiu

Axis Options - Output

PSTAT 130 - Summer 2017- William Qiu

Modifying Axis Labels
To place labels on the axes:
proc gplot data=ia.flight114;
where date between '02mar2001'd and
'08mar2001'd;

plot Boarded*Date / vaxis=100 to 200 by 25

ctext=blue;
symbol value=square i=join color=red width=2;
label Boarded='Passengers Boarded'
Date='Departure Date';
run;

ods html close;

PSTAT 130 - Summer 2017- William Qiu

Modifying Axis Labels

PSTAT 130 - Summer 2017- William Qiu

Produing a Scatterplot
 A scatterplot (usually) plots two continuous variables

proc gplot data=data1.admit;

plot weight*height;
run;
quit;

PSTAT 130 - Summer 2017- William Qiu

ScatterPlot

PSTAT 130 - Summer 2017- William Qiu

Adding Options
 You can modify the symbol, axis labels and axis tick marks
 You usually do not connect the dots in a scatterplot

proc gplot data=data1.admit;

plot weight*height;
symbol v=dot color=blue;
run;
quit;

PSTAT 130 - Summer 2017- William Qiu

Scatterplot with Regression
 You can also add a regression line, regression equation
and prediction confidence interval

proc gplot data=data1.admit;

plot weight*height / regeqn ;
symbol v=dot i=rlCLM95;
run;
quit;

PSTAT 130 - Summer 2017- William Qiu

Scatterplot with Regression

I = rl

I = rlclm95

regeqn
PSTAT 130 - Summer 2017- William Qiu
William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017-William Qiu
Lecture Outline
 Controlling When Records are Written to a SAS Dataset
 Creating Multiple Records from a Single Observations
(OUPUT)
 Writing to Multiple SAS Data Sets (OUTPUT)
 Creating a Single Record from Multiple Observations
(RETAIN)
 Creating a running total – RETAIN Statement and Sum
Expression
 Limiting the Variables Read from or Written to a SAS Data
Set (DROP= and KEEP=)
 Limiting the Number of Observations Read from a Data
Set (FIRSTOBS and OBS statements)
PSTAT 130 - Summer 2017-William Qiu
PSTAT 130 - Summer 2017-William Qiu
Automatic Output (Review)

data forecast;
2. Automatic set prog2.growth;
Return <additional SAS statements>;
run;

1. Automatic Output

PSTAT 130 - Summer 2017-William Qiu

The OUTPUT Statement
General Form

OUTPUT <SAS-data-set-1……SAS-data—set-n>;

PSTAT 130 - Summer 2017-William Qiu

OUTPUT Statement
 By default, SAS writes a data record to the output data set
when it reaches the RUN statement at the end of the
Data Step.
 If you include an OUTPUT statement in the Data Step
 SAS writes a record immediately
 and DOES NOT write automatically at the end of the data
step
 With OUTPUT you can
 Create multiple records from one observation
 Write to multiple output data sets
 Combine information from multiple observations into a
single record (when combined with RETAIN statement)
PSTAT 130 - Summer 2017-William Qiu
Creating Multiple Observations
PROCESS: Take each patient’s weight at admission and forecast
forward for the next five years
 Assume 5% weight gain per year
 Example: John Jones currently weighs 175 lbs. We project he will weigh
5% more each year:
Year Projected Weight
1 183.8
2 192.9
3 202.6
4 212.7
5 223.3

 Use the OUTPUT statement to write forecasted weight to the

output data set for each of the next five years

PSTAT 130 - Summer 2017-William Qiu

Input Data Set – ia.admit
Name Level Sex Age Height Weight

Murray, W HIGH M 27 72 168

Almers, C HIGH F 34 66 152
Bonaventure, T LOW F 31 61 123
Johnson, R MOD F 43 63 137
LaMance, K LOW M 51 71 158
Jones, M HIGH M 29 76 193
Reberson, P MOD F 32 67 151
King, E MOD M 35 70 173
Pitts, D LOW M 34 73 154
Eberhardt, S LOW F 49 64 172
Nunnelly, A HIGH F 44 66 140
Oberon, M LOW F 28 62 118
Peterson, V MOD M 30 69 147
Quigley, M HIGH F 40 69 163
Cameron, L MOD M 47 72 173
Underwood, K LOW M 60 71 191
Takahashi, Y MOD F 43 65 123

PSTAT 130 - Summer 2017-William Qiu

 Create Multiple Observations
data work.weightproj;
set ia.admit; Calculates 5%
weight increase for
increase = 1.05; Year 1
year = 1;
projweight = weight * increase;
output;
year = 2;
Writes Year 1
Projection to SAS projweight = projweight * increase;
Dataset output;
year = 3;
projweight = projweight * increase;
output;
year = 4;
projweight = projweight * increase;
output;
year = 5;
projweight = projweight * increase;
output;
run;
PSTAT 130 - Summer 2017-William Qiu
Log Output
NOTE: There were 21 observations read
from the data set IA.ADMIT.
NOTE: The data set WORK.WEIGHTPROJ has
105 observations and 12 variables.

PSTAT 130 - Summer 2017-William Qiu

Output Data Set – work.weightproj
Patient Projected
Name Year Weight

Murray, W 1 176.400
Murray, W 2 185.220
Murray, W 3 194.481
Murray, W 4 204.205
Murray, W 5 214.415
Almers, C 1 159.600
Almers, C 2 167.580
Almers, C 3 175.959
Almers, C 4 184.757
Almers, C 5 193.995
Bonaventure, T 1 129.150
Bonaventure, T 2 135.608
Bonaventure, T 3 142.388
Bonaventure, T 4 149.507
Bonaventure, T 5 156.983
PSTAT 130 - Summer 2017-William Qiu
Writing to Multiple SAS Data Sets
 GOAL: Use the ADMIT data set to create three separate
output data sets:
 Patients with Low activity level
 Patients with Moderate activity level
 Patients with High activity level
 This is the reverse of Concatenating (Appending) data sets

PSTAT 130 - Summer 2017-William Qiu

The DATA Statement (Review)
 General Form

DATA <data-set-name-1> <…data-set-name-n>;

Name of the FIRST Name of the SECOND

dataset being created by dataset being created by
this data step this data step

PSTAT 130 - Summer 2017-William Qiu

Input Data Set – ia.admit
Name Level Sex Age Height Weight

Murray, W HIGH M 27 72 168

Almers, C HIGH F 34 66 152
Bonaventure, T LOW F 31 61 123
Johnson, R MOD F 43 63 137
LaMance, K LOW M 51 71 158
Jones, M HIGH M 29 76 193
Reberson, P MOD F 32 67 151
King, E MOD M 35 70 173
Pitts, D LOW M 34 73 154
Eberhardt, S LOW F 49 64 172
Nunnelly, A HIGH F 44 66 140
Oberon, M LOW F 28 62 118
Peterson, V MOD M 30 69 147
Quigley, M HIGH F 40 69 163
Cameron, L MOD M 47 72 173
Underwood, K LOW M 60 71 191
Takahashi, Y MOD F 43 65 123

PSTAT 130 - Summer 2017-William Qiu

Create Multiple SAS Data Sets in a
Single DATA Step
data work.low work.mod work.high;
set ia.admit;
if Actlevel = 'LOW' then output low;
if Actlevel = 'MOD' then output mod;
if Actlevel = 'HIGH' then output high;
run;

PSTAT 130 - Summer 2017-William Qiu

Output Data Set – work.low
Name Level Sex Age Height Weight

Bonaventure, T LOW F 31 61 123

LaMance, K LOW M 51 71 158
Pitts, D LOW M 34 73 154
Eberhardt, S LOW F 49 64 172
Oberon, M LOW F 28 62 118
Underwood, K LOW M 60 71 191
Ivan, H LOW F 22 63 139

PSTAT 130 - Summer 2017-William Qiu

Creating One Record from Multiple Observations
Use the RETAIN and OUTPUT statements, and First./Last.
variables, to combine information from multiple data
records, and write only one “summary” record

 Example: Patients sign up for a six-month weight loss

program. They are weighed at the start of the program,
and then at their Month 3 and Month 6 visits. The data
set records the weight (and the date of the visit) on a
separate record for each visit.
 We need to calculate how much weight the subject has lost
since the last visit.
PSTAT 130 - Summer 2017-William Qiu
Weightloss Data Set
Obs name visit weight

1 john 01/01/08 210

2 john 04/04/08 199
3 john 07/03/08 183
4 mary 02/14/08 175
5 mary 05/16/08 167
6 mary 08/04/08 153
7 dave 09/23/08 223
8 dave 12/28/08 215
9 dave 03/24/09 206

PSTAT 130 - Summer 2017-William Qiu

RETAIN Statement
By default, SAS sets the values of all variables to missing at
the start of each pass through the Data Step. As a result,
you can’t save the values of a variable from one record to
the next.
Use the RETAIN statement to keep the value of a calculated
variable available for use with the next observation.
RETAIN variable-1 < . . . variable-n>;
 General Form:

PSTAT 130 - Summer 2017-William Qiu

Calculating variables using
consecutive records - RETAIN
data work.weightchange;
set weightloss;
weightchange = lastweight - weight;
lastweight = weight;
retain;
run;

PSTAT 130 - Summer 2017-William Qiu

Weight Change – Version 1 – RETAIN
Change
Subject Visit Visit in Last
Name Date Weight Weight Weight

john 01/01/08 210 . 210

john 04/04/08 199 11 199
john 07/03/08 183 16 183
mary 02/14/08 175 8 175
mary 05/16/08 167 8 167
mary 08/04/08 153 14 153
dave 09/23/08 223 -70 223
dave 12/28/08 215 8 215
dave 03/24/09 206 9 206

PSTAT 130 - Summer 2017-William Qiu

First.ByVariable & Last.ByVariable
When PROC SORT is used to sort a data set, it creates two
variables that can be used to identify the First record in
the By Group and the Last record in the By Group.

 General Form
First.BY-variable
Last.BY-variable

PSTAT 130 - Summer 2017-William Qiu

Calculating Weight Change
data weightchange;
set weightloss;
by name visit;
if first.name then do;
baseweight=weight;
weightchange = .;
end;
else weightchange = baseweight - weight;
retain;
run;

PSTAT 130 - Summer 2017-William Qiu

Weight Change – Version 2
RETAIN with First.By
Change
Subject Visit Visit Baseline in
Name Date Weight Weight Weight

dave 09/23/08 223 223 .

dave 12/28/08 215 223 8
dave 03/24/09 206 223 17
john 01/01/08 210 210 .
john 04/04/08 199 210 11
john 07/03/08 183 210 27
mary 02/14/08 175 175 .
mary 05/16/08 167 175 8
mary 08/04/08 153 175 22

First.Name
Last.Name
PSTAT 130 - Summer 2017-William Qiu
Outputting Only Summary Records
data weightchange;
set weightloss;
by name visit;
if first.name then do;
baseweight=weight;
weightchange = .;
end;
if last.name then do;
weightchange = baseweight - weight;
output;
end;
retain;
run;

PSTAT 130 - Summer 2017-William Qiu

Weight Change – Summary
Change
Subject Visit Visit Baseline in
Name Date Weight Weight Weight

dave 03/24/09 206 223 17

john 07/03/08 183 210 27
mary 08/04/08 153 175 22

PSTAT 130 - Summer 2017-William Qiu

Controlling Variable Output
By default, the SAS System writes all variables from every
input data set to every output data set. In the DATA step,
the DROP and KEEP statements can be used to control
which variables are written to output data sets.

PSTAT 130 - Summer 2017-William Qiu

The DROP and KEEP Statements (Review)

PSTAT 130 - Summer 2017-William Qiu

The DROP and KEEP Statements (Review)

PSTAT 130 - Summer 2017-William Qiu

DROP= and KEEP= Data Set Options
 DROP=
SAS-data-set(DROP=variable-1 variable-2 …variable-n)

 KEEP=
SAS-data-set(KEEP=variable-1 variable-2 …variable-n)

PSTAT 130 - Summer 2017-William Qiu

DROP= and KEEP= Data Set Options

PSTAT 130 - Summer 2017-William Qiu

 Data2.Military – Partial Proc Print
Type Code City State Country Airport

Air Force NSF Camp Springs MD USA Andrews Air Force Base
Army EDG Edgewood Arsenal MD USA Weide Army Air Field
Army FME Fort Meade (Odenton) MD USA Tipton Army Air Field
Naval NHZ Brunswick ME USA Brunswick Naval Air
Air Force LIZ Limestone ME USA Loring Air Force Base
Air Force SAW Gwinn/Marquette MI USA K. I. Sawyer Air Force
Marine 76G Marine City MI USA Marine City Airport
Naval NFB Mount Clemens MI USA Detroit Naval Air
Air Force SZL Knob Noster MO USA Whiteman Air Force Base
Air Force BIX Biloxi MS USA Keesler Air Force Base
Air Force CBM Columbus MS USA Columbus Air Force Base
Naval NJW Moscow MS USA Joe Williams Naval
Air Force GFA Great Falls MT USA Malmstrom Air Force Base
Air Force POB Fayetteville NC USA Pope Air Force Base
Army FBG Fort Bragg NC USA Simmons Army Air Field
Air Force GSB Goldsboro NC USA Seymour Johnson Air Force
Army HFF Hoffmann/Camp Mackal NC USA Mackall Army Air Field

PSTAT 130 - Summer 2017-William Qiu

Controlling Variable Output
data airforce(drop=Code Type)
army(drop=City State Country Type)
navy(drop=Type)
marines;
set prog2.military;
if Type eq 'Air Force' then
output airforce;
else if Type eq 'Army' then
output army;
else if Type eq 'Naval' then
output navy;
else if Type eq 'Marine' then
output marines;
run;

PSTAT 130 - Summer 2017-William Qiu

Controlling Variable Output

PSTAT 130 - Summer 2017-William Qiu

Controlling Variable Output

data army(keep=Code Airport);

set prog2.military(drop=City State
Country);
if Type eq 'Army' then output;
run;

PSTAT 130 - Summer 2017-William Qiu

Compiling the Data Step

PSTAT 130 - Summer 2017-William Qiu

Compiling the Data Step

PSTAT 130 - Summer 2017-William Qiu

Controlling Which Observations are Read
The OBS= Data Set Option determines how many
observations are read from the input data set
 General Form:
SAS-data-set(OBS=n)

 Example:
data army;
set prog2.military(obs=25);
if Type eq 'Army' then output;
run;

PSTAT 130 - Summer 2017-William Qiu

Controlling Which Observations are Read

PSTAT 130 - Summer 2017-William Qiu

 Controlling Which Observations are Read
The FIRSTOBS= Data Set Option allows you to skip records
at the top of the input file. Can be useful to eliminate
“header” information.
 General Form:
SAS-data-set(FIRSTOBS=n)

 Example (starts reading at 2nd record):

PSTAT 130 - Summer 2017-William Qiu

Controlling Which Observations are Read

PSTAT 130 - Summer 2017-William Qiu

William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017- William Qiu
Lecture Outline
 Learn another way to create a running total
 Sum statement
 Clarify First.Obs and Last.Obs variables
 With Single and Multiple BY variables
 Writing to external files
 Creating a comma-separated (or Excel) data file - ODS CSVALL
 Creating a custom report – DATA _NULL_
 Reading Non-standard Data files
 Delimiters other than spaces
 Mixed Format Records
 Multiple records per line

PSTAT 130 - Summer 2017- William Qiu

The SUM Statement
The SUM statement (General Form)
Variable + expression;

 creates the variable on the left side of the plus sign if it

does not already exist
 initializes the variable to zero before the first iteration of
the DATA step
 automatically retains the variable
 adds the value of the expression to the variable at
execution
 ignores missing values.
 is an alternative to using the RETAIN statement
PSTAT 130 - Summer 2017- William Qiu
Accumulating Totals
data mnthtot2;
set prog2.daysales2;
Mth2Dte+SaleAmt;
run;

SaleDate SaleAmt Mth2Dte

01APR2001 498.49 498.49

02APR2001 946.50 1444.99
03APR2001 994.97 2439.96
04APR2001 564.59 3004.55
05APR2001 783.01 3787.56
Accumulating 06APR2001 228.82 4016.38
Variable 07APR2001 930.57 4946.95
08APR2001 211.47 5158.42
09APR2001 156.23 5314.65

PSTAT 130 - Summer 2017- William Qiu

Accumulating totals for BY Groups
After sorting the data on the BY variables, there is a three-
step process for accumulating totals.
1. Set the Accumulator Variable to zero at the start of each
BY group.
2. Increment the accumulating variable with a sum
statement (automatically RETAINs).
3. Output only the last observation of each BY group.

PSTAT 130 - Summer 2017- William Qiu

First. & Last. Example

PSTAT 130 - Summer 2017- William Qiu

data divsals(keep=Div DivSal);
set salsort;
by Div;
if First.Div then DivSal=0;
DivSal+Salary;
if Last.Div;
run;

PSTAT 130 - Summer 2017- William Qiu

Sorting using Multiple BY variables
proc sort data=prog2.regsals out=regsort;
by Region Div;
run;

PSTAT 130 - Summer 2017- William Qiu

Writing Data to an Excel file
 You can use the ODS HTML and PROC PRINT statements
to convert a SAS data set into an Excel file
ods html file='Admit.xls';
title;
proc print data=ia.admit noobs;
run;
ods html close;

 Note: SAS will warn you that this file is not of .xls type.
Just ignore this warning.

PSTAT 130 - Summer 2017- William Qiu

Writing a Comma-Separated File
 Use the ODS CSVALL statement to convert a SAS data set
to a comma-separated values (CSV) file
 Most other programs can read in a CSV file, including
Excel
ods csvall file='admit.csv';
title;
proc print data=ia.admit noobs ;
run;
ods csvall close;

 You can use titles, footnotes, labels and formats to

change the appearance of the data

PSTAT 130 - Summer 2017- William Qiu

CSVALL Output
"ID","Name","Sex","Age","Date","Height","Weight","ActLevel","Fee"
2458,"Murray, W","M",27,1,72,168,"HIGH",85.20
2462,"Almers, C","F",34,3,66,152,"HIGH",124.80
2501,"Bonaventure, T","F",31,17,61,123,"LOW",149.75
2523,"Johnson, R","F",43,31,63,137,"MOD",149.75
2539,"LaMance, K","M",51,4,71,158,"LOW",124.80
2544,"Jones, M","M",29,6,76,193,"HIGH",124.80
2552,"Reberson, P","F",32,9,67,151,"MOD",149.75
2555,"King, E","M",35,13,70,173,"MOD",149.75
2563,"Pitts, D","M",34,22,73,154,"LOW",124.80
2568,"Eberhardt, S","F",49,27,64,172,"LOW",124.80
2571,"Nunnelly, A","F",44,19,66,140,"HIGH",149.75
2572,"Oberon, M","F",28,17,62,118,"LOW",85.20
2574,"Peterson, V","M",30,6,69,147,"MOD",149.75
2575,"Quigley, M","F",40,8,69,163,"HIGH",124.80
2578,"Cameron, L","M",47,5,72,173,"MOD",124.80
2579,"Underwood, K","M",60,22,71,191,"LOW",149.75
2584,"Takahashi, Y","F",43,29,65,123,"MOD",124.80
2586,"Derber, B","M",25,23,75,188,"HIGH",85.20
2588,"Ivan, H","F",22,20,63,139,"LOW",85.20
2589,"Wilcox, E","F",41,16,67,141,"HIGH",149.75
2595,"Warren, C","M",54,7,71,183,"MOD",149.75

PSTAT 130 - Summer 2017- William Qiu

More about the Data Step

 Note: INFILE vs. FILE, INFORMAT vs. FORMAT, INPUT vs. PUT
 One controls data coming into SAS, the other controls data leaving SAS

PSTAT 130 - Summer 2017- William Qiu

Creating Custom Reports
 You can use the PUT statement with the DATA _NULL_
method to write output in any format you choose.
 DATA _NULL_ uses the Data Step, but does NOT create a
data set.
 Instead, it writes output to a specified file using PUT
statements
 The PUT statements control the exact location and format
of information in the Output file

PSTAT 130 - Summer 2017- William Qiu

DATA _NULL_ Program and Output
data _null_;
set ia.admit;
if Sex = 'M' then Salutation = 'Mr.'; Else Salutation = 'Ms.';
file 'admitreport.doc' PRINT;
title;
put @5 'Dear ' Salutation ' ' Name ':'
//@5 'Your weight at admission was ' Weight 3. ' pounds.'
/@5 'Your height at admission was ' Height 2. ' inches.'
//@10 'Thank you for your ' fee dollar8.2 ' payment!';
put _page_;
run;

Dear Mr. Murray, W :

Your weight at admission was 168 pounds.

Your height at admission was 72 inches.

Thank you for your $85.20 payment!

PSTAT 130 - Summer 2017- William Qiu

Colon Modifier – “Short” Variables
Use the Colon Modifier to read each value only as far as the
next delimiter
 Allows you to use Informats with List input, but handle
non-standard data values (i.e., values that do not match
the format)
data airplanes;
infile ‘airdata.txt';
input ID $
InService : date9.
PassCap CargoCap;
run;

PSTAT 130 - Summer 2017- William Qiu

Colon Modifier – Character Example
The default length for character variables is 8. Use the
Colon Modifier to read character values longer than 8
characters. The ‘:’ tells SAS to read in that variable until it
reaches a space:
data students;
input Name : $20. Gender : $6. Age 2.;
datalines;
elizabeth female 23
jonathon male 22
zack male 19
;
run;

PSTAT 130 - Summer 2017- William Qiu

INFILE Statement Options

PSTAT 130 - Summer 2017- William Qiu

Using Other Delimiters
 A delimiter is used to separate values of adjacent
variables
 In SAS, the default delimiter is a “space”, but many other
programs use commas or tabs
Space delimited Comma delimited
John 22 M John,22,M
Betty 19 F vs. Betty,19,F
Dave 18 M Dave,18,M

PSTAT 130 - Summer 2017- William Qiu

The DLM= option
 Use the DLM= option to specify a delimiter
 The delimiter can be any character
 You can use the DLM option with
 INFILE statements (when reading in data) or
 with FILE statements (when writing out data)
data new;
infile 'students.dat' dlm=',';
input Name $ Age Gender $;
run;

Mary,21,Female
John,22,Male
David,18,Male
PSTAT 130 - Summer 2017- William Qiu
Missing Data at the End of a Row
By default, when there is missing data at the end of a row,
1. SAS loads the next record to finish the observation
2. a note is written to the log
3. SAS loads a new record at the top of the DATA step and
continues processing.

PSTAT 130 - Summer 2017- William Qiu

Missing Data at the End of a Row

PSTAT 130 - Summer 2017- William Qiu

Missing Data at the End of a Row

PSTAT 130 - Summer 2017- William Qiu

Solution – MISSOVER option
 The MISSOVER option tells SAS NOT to continue reading
from the next line if not all variables have values
data airplanes3;
infile ‘airdata.txt’dlm=',' missover;
input ID $
InService : date9.
PassCap CargoCap;
run;

In Pass Carg
ID Service Cap Cap

50001 10627 132 .

50002 10907 152 540
50003 11617 90 530
50004 12088 172 .
50005 12228 170 510
50006 12772 180 520

PSTAT 130 - Summer 2017- William Qiu

The DSD Option
The DSD option:
 sets the default delimiter to a comma
 treats consecutive delimiters as missing values
 enables SAS to read values with embedded delimiters if
the value is surrounded by double quotes.

PSTAT 130 - Summer 2017- William Qiu

Mixed Record Types
What if not all records have the same format?
Data file

Desired Output

PSTAT 130 - Summer 2017- William Qiu

The Trailing @ modifier
 The Trailing @ modifier tells SAS to hold the current input
line for further processing:
input SalesID $ Location $ @;
if location='USA' then
input SaleDate : mmddyy10.
Amount;
else if Location='EUR' then
input SaleDate : date9.
Amount : commax8.;

PSTAT 130 - Summer 2017- William Qiu

Mixed Record Types

PSTAT 130 - Summer 2017- William Qiu

Multiple Observations Per Line

PSTAT 130 - Summer 2017- William Qiu

The Double Trailing @@ modifier

PSTAT 130 - Summer 2017- William Qiu

Trailing @ versus Double Trailing @@

PSTAT 130 - Summer 2017- William Qiu

William Qiu
Department of Statistics and Applied Probability
UCSB
PSTAT 130 - Summer 2017 - William Qiu
Lecture Outline
 Variable Lists
 Using SAS functions
 Parse text data
 Truncate numeric data
 Converting data
 numeric to character
 character to numeric
 Do loops
 Non-iterative
 Iterative
 Condition

PSTAT 130 - Summer 2017 - William Qiu

SAS Variable Lists
SAS enables you to use the following variable lists:
 numbered range lists
 name range lists
 name prefix lists
 special SAS name lists.

PSTAT 130 - Summer 2017 - William Qiu

Numbered Range List
 Allows you to refer to a set variables that all start with the
same variable name, and each end with a number
 Example:
Input Salesman $ Week1 Week2 Week3 Week4...Week52;

 A range list refers to all the weeks as follows:

Proc Print;
Var Week1-Week52;
Run;

PSTAT 130 - Summer 2017 - William Qiu

Name Range List
 Allows you to refer to a series of variables that appear in
consecutive order in the dataset
 Example:
Input Salesman $ Mon Tues Wed Thur Fri Sat Sun Region;

 A range list refers to all the weeks as follows:

Proc Print;
Var Mon--Sun;
Run;
 Note: Salesman-Numeric-Region includes only Numeric variables in
the range, and Salesman-Character-Region includes on character
variables in the range

PSTAT 130 - Summer 2017 - William Qiu

Name Prefix List
 Some SAS functions and statements allow you to refer to
all variables that begin with a specified character string:
 Example:
sum(of SALES:)

tells SAS to calculate the sum of all the variables that

begin with "SALES," such as SALES_JAN, SALES_FEB, and
SALES_MAR.

PSTAT 130 - Summer 2017 - William Qiu

Special SAS Names List
 Special SAS name lists include
_NUMERIC_ specifies all numeric variables that are already
defined in the current DATA step.
_CHARACTER_ specifies all character variables that are currently
defined in the current DATA step.
_ALL_ specifies all variables that are currently defined in the
current DATA step.

PSTAT 130 - Summer 2017 - William Qiu

SAS Functions
A SAS function is often categorized by the type of data manipulation
performed:
 truncation
 character
 date and time
 mathematical
 trigonometric
 special
 sample statistics
 arithmetic
 financial
 random number
 state and ZIP code.

PSTAT 130 - Summer 2017 - William Qiu

Example: Making Mailing Labels
The data2.freqflyers data set contains information about
frequent flyers.

Use this data set to create another data set suitable for
mailing labels.

PSTAT 130 - Summer 2017 - William Qiu

The LENGTH Function
 The LENGTH function returns the number of characters in
a string
NewVar = LENGTH(string);

 Example
 LENGTH(“SMITH, JOHN”) = 11

PSTAT 130 - Summer 2017 - William Qiu

The INDEX Function
 The INDEX function returns the position of specific
character (or characters) within a string
NewVar = INDEX(string,target);

 Example
 INDEX(‘SMITH-JOHN’,’-’) = 6

 Returns ZERO if target isn’t in string

PSTAT 130 - Summer 2017 - William Qiu

The SUBSTR Function (Right Side)
 The SUBSTR Function extracts a portion of a character
variable:
NewVar=SUBSTR(string,start<,length>);

 Example:
 SUBSTR(“PSTAT130”,6,3) = “130”

PSTAT 130 - Summer 2017 - William Qiu

Parsing a Text String
 How can we turn “SMITH, JOHN” into “JOHN SMITH“?
 Find the location of the comma
 Last Name = text before the comma
 First Name = Text after the comma

PSTAT 130 - Summer 2017 - William Qiu

 Putting it all together
data mailabels;
input name $25.;
name_len = length(name);
comma_pos = index(name,',');
last_name = substr(name,1,comma_pos-1);
first_name = substr(name,comma_pos+2,name_len-comma_pos);
datalines;
Smith, John
Johnson, Davy
Quincy, Elizabeth
;
run;

PSTAT 130 - Summer 2017 - William Qiu

Name Example

123456789012345
Smith, John

Length= 11
Comma_pos = 6

Last_name = First_name =
SUBSTR(1,5) SUBSTR(8,3)

PSTAT 130 - Summer 2017 - William Qiu

Results
name name_len comma_pos last_name first_name

Smith, John 11 6 Smith John

Johnson, Davy 13 8 Johnson Davy

Quincy, Elizabeth 17 7 Quincy Elizabeth

PSTAT 130 - Summer 2017 - William Qiu

The SCAN Function
 The SCAN function “parses” a character string into a set
of “words” using a delimiter.
NewVar=SCAN(string,n<,delimiters>)

First “word”
 Example:
 SCAN(“Smith, John”, 1) = “Smith”
 SCAN(“Smith, John”, 2) = “John”

Second “word”

PSTAT 130 - Summer 2017 - William Qiu

When the SCAN function is used,
 The default delimiters are:
blank . < ( + | & ! $ * ) ; ¬ - / , % | ¢
 delimiters before the first word have no effect
 any character or set of characters can serve as delimiters
 two or more contiguous delimiters are treated as a single
delimiter
 a missing value is returned if there are fewer than n
words in string
 if n is negative, SCAN selects the word in the character
string starting from the end of string.

PSTAT 130 - Summer 2017 - William Qiu

Concatenation Operator
 Use the || operator to “concatenate” or join two strings
together

 Examples:
 “John” || “Smith” = “JohnSmith”
 “John” || “ “ || “Smith” = “John Smith”

PSTAT 130 - Summer 2017 - William Qiu

The TRIM Function
 The TRIM function removes any trailing blank spaces

 Example:
 TRIM(“JOHN “) = “JOHN”

PSTAT 130 - Summer 2017 - William Qiu

A Better Mailing Label Program
data test;
input name $25.;
last_name = scan(name,1);
first_name = scan(name,2);
datalines;
Smith, John
Johnson, Davy
Quincy, Elizabeth
;
run;

name last_name first_name

Smith, John Smith John
Johnson, Davy Johnson Davy
Quincy, Elizabeth Quincy Elizabeth

PSTAT 130 - Summer 2017 - William Qiu

A Search Application
The data2.ffhistory data set contains information about the
history of each frequent flyer.
 This history information consists of
 each membership level the flyer has attained (bronze,
silver, or gold)
 the year the flier attained each level.
Create a report that shows all frequent flyers who have
attained silver membership status and the year each of
them became silver members.

PSTAT 130 - Summer 2017 - William Qiu

A Search Application

 How would you

 Parse the membership levels?
 Parse the year each level was obtained?
 Select flyers that have achieved Silver status?

PSTAT 130 - Summer 2017 - William Qiu

Numeric Truncation Functions
Selected functions that truncate numeric values include:
 ROUND function
 CEIL function
 FLOOR function
 INT function.

PSTAT 130 - Summer 2017 - William Qiu

The ROUND Function
 The ROUND function performs a traditional Round
Up/Round Down operation:
NewVar=ROUND(argument<,round-off-unit>);
 Examples:

PSTAT 130 - Summer 2017 - William Qiu

The CEIL Function
 The CEIL Function performs a Round Up operation only:
NewVar = CEIL(argument);

 Note: CEIL(4) = 4

PSTAT 130 - Summer 2017 - William Qiu

The FLOOR Function
 The FLOOR Function performs a Round Down operation
only:
NewVar = FLOOR(argument);

 Note: FLOOR(4) = 4

PSTAT 130 - Summer 2017 - William Qiu

The INT Function
 The INT function removes any decimals and returns the
next smaller integer:
NewVar = INT(argument);
 Examples:
 INT(3.2) = 3
 INT(-4.8) = -4
 For positive numbers, INT = FLOOR
 For negative numbers, INT = CEIL

PSTAT 130 - Summer 2017 - William Qiu

Data Conversion
In many applications, you may need to convert one data
type to another.
 You may need to read in digits in character form and
convert them to a numeric value.
 You may need to read in a numeric value and write it out
as a character string.
You can convert data types
 implicitly by allowing the SAS System to do it for you
 explicitly with these functions:
 INPUT character-to-numeric conversion
 PUT numeric-to-character conversion.

PSTAT 130 - Summer 2017 - William Qiu

Automatic Character-to-Numeric Conversion
The data2.salary1 data set contains a character variable
Grosspay. Compute a 10 percent bonus for each
employee.
What will happen when the character values of Grosspay
are used in an arithmetic expression?

PSTAT 130 - Summer 2017 - William Qiu

Automatic Character-to-Numeric Conversion
SAS automatically converts a character value to a numeric
value when the character value is used in a numeric
context, such as
 assignment to a numeric variable
 an arithmetic operation
 logical comparison with a numeric value
 a function that takes numeric arguments.

PSTAT 130 - Summer 2017 - William Qiu

The INPUT Function
 The INPUT function uses a SAS format (informat) to
convert a character string into a number:
NumVar=INPUT(source,informat);

PSTAT 130 - Summer 2017 - William Qiu

Failed Character-to-Numeric Conversion
 This approach will not work because Grosspay has non-
numeric characters (commas).

PSTAT 130 - Summer 2017 - William Qiu

Explicit Character-to-Numeric Conversion

 Note: You cannot convert data by assigning the converted

variable value to a variable with the same name.
GrossPay=input(GrossPay,comma6.);
PSTAT 130 - Summer 2017 - William Qiu
Automatic Numeric-to-Character Conversion
SAS automatically converts a numeric value to a character
value when the numeric value is used in a character
context, such as
 assignment to a character variable
 a concatenation operation
 a function that accepts character arguments.

PSTAT 130 - Summer 2017 - William Qiu

The PUT Function
 The PUT function uses a SAS format to convert a number
into a character string:
CharVar=PUT(source,format);

 Example
 Areacode = 805
 AreaChar = PUT(Areacode,3.) = “805”

PSTAT 130 - Summer 2017 - William Qiu

PUT vs. INPUT Functions
 These are NOT the same as the PUT and INPUT
statements used in a Data Step
 Functions
 The INPUT function converts character data into numeric
data
 The PUT function converts numeric data into character data
 Statements in Data Step
 The INPUT statement controls how data are read into a data
set
 The PUT statement controls how data are written out of a
data step
PSTAT 130 - Summer 2017 - William Qiu
Do Loop Processing
There are four kinds of DO Loops in SAS:
 DO-END
 executes statements as a unit, usually as a part of IF-THEN/ELSE
statements.
 Iterative DO
 executes a group of statements repetitively based on the value of an
index variable.
 DO WHILE
 executes a group of statements repetitively as long as the condition
that you specify remains true. The condition is checked before each
iteration of the loop.
 DO UNTIL
 executes a group of statements repetitively until the condition that you
specify is true. The condition is checked after each iteration of the loop.

PSTAT 130 - Summer 2017 - William Qiu

DO-END
The simple IF-THEN statement does NOT require a DO-END
statement:
IF sex = ‘Male’ THEN Abbreviation = ‘Mr.’;

BUT if you want to do more than one thing, you need a DO-
END statement
IF sex = ‘Male’ THEN
DO;
Abbreviation = ‘Mr.’;
Salutation = ‘Sir’;
END;

PSTAT 130 - Summer 2017 - William Qiu

Performing Repetitive Calculations
On January 1 of each year, $5,000 is invested in an account.
Determine the value of the account after three years
based on a constant annual interest rate of 7.5 percent.
data invest;
do Year=2001 to 2003;
Capital+5000;
Capital+(Capital*.075);
end;
run;

Year Capital

2004 17364.61

PSTAT 130 - Summer 2017 - William Qiu

Performing Repetitive Calculations
Generate the list of yearly increases
data invest;
do Year=2001 to 2003;
Capital+5000;
Capital+(Capital*.075);
output;
end;
run;

Year Capital
2001 5375.00
2002 11153.13
2003 17364.61

PSTAT 130 - Summer 2017 - William Qiu

Performing Repetitive Calculations
Use the iterative DO to repeat an operation a fixed number
of times:
data iterate;
Summation = 0;
Factorial = 1;
n = 10;
do i = 1 to n;
Summation = Summation + i;
Factorial = Factorial * i;
output;
retain;
end;
run;

PSTAT 130 - Summer 2017 - William Qiu

The Iterative DO Statement- Output
Obs i Summation Factorial

1 1 1 1
2 2 3 2
3 3 6 6
4 4 10 24
5 5 15 120
6 6 21 720
7 7 28 5040
8 8 36 40320
9 9 45 362880
10 10 55 3628800

PSTAT 130 - Summer 2017 - William Qiu

A Forecasting Application (Review)
data forecast;
set data2.growth(rename=(NumEmps=NewTotal));
Increase = .1;
Year=1;
NewTotal=NewTotal*(1+Increase);
output;
Year=2;
NewTotal=NewTotal*(1+Increase);
output;
Year=3;
NewTotal=NewTotal*(1+Increase);
output;
data forecast;
run;
set prog2.growth(rename=(NumEmps=NewTotal));
Increase = .1;
do Year=1 to 3;
NewTotal=NewTotal*(1+Increase);
output;
end;
run;

PSTAT 130 - Summer 2017 - William Qiu

The Iterative DO Statement
 The Index Variable can start or stop on any number
do i = 5 to 20;
<SAS Statements>
end;

 You can iterate in multiples using the BY option

data odd numbers;
do i = 1 to 100 by 2;
output;
end;
run;

 Index variables can be User-Defined lists:

do Day = 'Mon','Tue','Wed','Thu','Fri','Sat','Sun';
output;
end;

PSTAT 130 - Summer 2017 - William Qiu

The Iterative DO Statement
How many times will each DO loop execute?

PSTAT 130 - Summer 2017 - William Qiu

DO WHILE and DO UNIT Loops
A DO-WHILE Loop executes a group of statements repetitively as
long as the condition that you specify remains true. The
condition is checked before each iteration of the loop.
A DO-UNTIL Loop executes a group of statements repetitively
until the condition that you specify is true. The condition is
checked after each iteration of the loop.

PSTAT 130 - Summer 2017 - William Qiu

Conditional Iterative Processing
Determine the number of years it would take for an account
to exceed $1,000,000 if $5,000 is invested annually at 7.5
percent interest.
data invest;
do until(Capital>1000000);
Year+1;
Capital+5000;
Capital+(Capital*.075);
end;
run;

PSTAT 130 - Summer 2017 - William Qiu

Iterative DO with Conditional Clause
Determine the return of the account again.
Stop the loop if 25 years is reached or more than $250,000
is accumulated.

PSTAT 130 - Summer 2017 - William Qiu