BIODATA SINGKAT

: Prof. DR. SAMSUBAR SALEH. M.Soc.Sc

: DOSEN FEB-UGM

: GURU BESAR ILMU EKONOMI IV/c

: JL WELING II / 39 A. DEPOK-SLEMAN

: EKONOMI-PUBLIK
STATISTIKA TERAPAN
TEORI EKONOMI MAKRO/MIKRO
PENGALAMAN KERJA : DOSEN FEB-UGM
: ASSISTEN WADEK III & SEK-JURUSAN IE.
: PENGELOLA S2/S3 FEB-UGM
: SUPERVISOR BSNP DIKNAS
: PENGELOLA PENELITIAN EKONOMI
: KONSULTAN PERENC/PEMB DIY
: KONSULTAN & FASILITATOR KEU-DA
HP: 0811253224.
Email :[email protected]

NAMA
PEKERJAAN
JABATAN
ALAMAT
BIDANG MINAT

Text-Books

1. Lind & Marchal; Statistics for Business

2. Wonnacott : Statistics
3. Samsubar Saleh: Statistik Induktif

PROBABILITY
PROBABILITY IS A NUMERICAL MEASURE OF THE
LIKELIHOOD OF AN EVENT OR MORE OCCURRING or
CHANCE AN EVENT OR MORE WILL OCCUR AS A RESULT
OF AN EXPERIMENT.
THE CONCEPTS OF PROBABILITY ARE SPECIALLY
RELEVANT IN BUSINESS. BUSINESSMEN MAKE DECISIONS
IN THE FACE OF UNCERTAINTY DAY BY DAY. FOR
EXAMPLE:
1. WHEN COMPANY LAUNCHES A NEW PRODUCT.
2.THE LIKELIHOOD OF A NEW INVESTMENT PROJECT
BEING PROFITABLE MAY BE UNKNOWN. THEREFORE,
DECISION MAKING HAS TO BE BASED ON SOME
ASSESSMENT OF THE PROBABILITY OF POSSIBLE
OUTCOMES OCCURRING.
3.ENGINEERING INSTALATION DESIGN.

EXPERIMENT, OUTCOMES AND EVENT

EACH EXPERIMENT WILL HAVE TWO OR MORE POSSIBLE

RESULTS.
OUTCOME; A PARTICULAR RESULT OF AN EXPERIMENT.
EVENT ; A COLLECTION OF ONE OR MORE OUTCOMES OF AN
EXPERIMENT.

EXPERIMENT:
ALL POSSIBLE
OUTCOMES

SOME POSSIBLE
EVENTS

TOSS 2 COINS
HH
HT
TH
TT
TAIL WILL APPEAR
AT LEAST ONCE

ROLL 2 DICE
THERE ARE 36
POSSIBLE OUTCOMES

TOTAL NUMBER OF
BOTH DICE = 2 OR 10

THE MULTIPLICATION RULE OF PROBABILITY

INDEPENDENT EVENTS: TWO EVENTS ARE
INDEPENDENT WHEN THE OCCURRENCE OF ONE
EVENT HAS NO EFFECT ON THE PROBABILITY OF
OCCURRENCE OF THE OTHER EVENT.
P ( A and B ) = P ( A ) . P ( B )
1.

2. DEPENDENT EVENT: TWO EVENTS ARE DEPENDENT

WHEN THE OCCURRENCE OF ONE EVENT AFFECT THE
PROBABILITY OF OCCURRENCE OTHER EVENT.
P ( A and B ) = P ( A ) . P ( B/A )

Example

a.
b.
c.

d.
e.

A BOX CONTAINS OF 10 RED BALLS, 6 WHITE

BALLS AND 4 BLUE BALLS. ONE BALL IS
SELECTED AT RANDOM 3 TIMES. WHAT IS
THE PROBABILITY of OBTAINING:
All balls are white ( with replacement ).
All balls are white ( without replacement ).
The 1st selection is red, the 2nd selection is red and the 3rd
selection is blue ( without replacement ).
There are 2 red balls and 1 blue ball ( without replacement )
There are 1 red, 1 white and 1 blue ball ( Without R).

THE RULES OF PROBABILITY

THE ADDITION RULE OF PROBABILITY:
1. FOR MUTUALLY EXCLUSIVE EVENTS.
P ( A or B ) = P ( A ) + P ( B )
EVENTS A & B CANT OCCUR SIMULTANEOUSLY
OR AT THE SAME TIME.
2. FOR NON-MUTUALLY EXCLUSIVE EVENTS:
P ( A or B ) = P ( A ) + P ( B ) P ( A and B )
A and B CAN OCCUR SIMULTANEOUSLY.

CONDITIONAL PROBABILITY

THE PROBABILITY OF A PARTICULAR

EVENT OCCURING, GIVEN THAT
ANOTHER EVENT HAS ALREADY
OCCURRED.
IF, P ( A and B ) = P ( A ) . P ( B/A ), THUS;
P ( A and B )
P ( B/A ) = ------------------------P( A)

EXAMPLE
SCORES : A
CLASSES

X
Y
Z

10
13
8

15
16
11

8
7
6

2
4
5

CALCULATE THE PROBABILITY:

1. P ( A or B or F )
4. P ( Z or A )

2. P ( X or C )
5. P ( C/X )

3. P ( Y or Z )
6. P ( Z/F )

Example

In Lab: A, there are 10 units Acer computer, 5 units Compact

and 15 units Toshiba. In lab: B, there are 15 units Acer
computers, 10 unit Compact and 10 units Toshiba. In lab: C,
there are 10 units Acer, 10 units Compact and 10 units
Toshiba.
A. If each lab is selected one computer randomly, what is the
probability that all of the computers are Toshiba.
B. If one computer is selected at random from one lab, what is
the probability that the computer is Acer or Compact.

BAYESS RULE FOR CONDITIONAL PROBABILITY

P ( Ai ) P ( B/Ai )
P ( Ai/B ) = ---------------------------------------------------------P ( A1) P ( B/A1 ) + -------+ P ( An ) P ( B/An )
EXAMPLE 1:
A car repair firm employs four paint sprayers, Tomy, Dicky , John and Harry.
Tomy is responsible for painting 20 % of all the cars produced, Dicky for 25 %,
John for 15 % and Harry for the remaining 40 %. On the basis of frequent
quality inspections it is discovered that, on average, 15 % of the car sprayed by
Tomy fall below the minimum standard, while for Dicky is 8 %, for John is 5%
and for Harry 10 %. If a car is selected at random is judged to be substandard.
a.
What is the probability that it was sprayed by Harry?
b. If a car is selected at random is judged to be standard, what is the probability
that it was sprayed by Tomy or John?

Example:2

The number of students from class A = 35, class B

= 30, class C = 40 and class D = 25. Based on the
examination record, it was found that 10% of
students from class A was fail, class B = 5%, class
C = 15 % and class D = 20 %. If one student was
selected at random was indicated that he did well
in the exam. What is the probability that he is
from class B or D.

Mathematical Expectation

Finding the expected value of the Mean, Variance and

Standard Deviation of a probability distribution:

Mean = Xi . P ( Xi ), where Xi is the value of variable X, and

P (Xi) is the probability of X.
Variance = ( Xi - )2 P ( Xi) or Xi2 P ( Xi ) - 2

Std dev = variance

Example

A family has five children. Please calculate the

expected value and standard deviation of having male
children.
Step1 : find out the value of Xi ( male children ).
Step 2: find out the probability of having Xi.
Xi
: ?
P( Xi ): ?

Example
The quantity of X demanded probability
200 units
0.15
225
0.20
250
0.30
275
0.25
300
0.10
a. What is the mean of quantity of X demanded.
b. What is the variance of the distribution?
c. When the price of X = $ 10 and AC = $ 8
What is the expected value of the profit?

Probability distributions
1. Discrete probability distribution:
Binomial distribution:
a.
In each trials only two outcomes are possible:
success or failure, good or defect, pass or fail, girl
or boy, odd number or even number.
b.
Each trial is independent to each other.
c.
Probability of an event is assumed to remain
constant over all trials ( n trials ).

Binomial Distribution

The probability of obtaining the outcome success denoted by and the outcome failure
denoted q or (1 ).
If we are interested in the number of success or failure ( x ) occuring in ( n ) trials, then, the
probability binomial distribution is given by the formula:
n

nx

x
Example:
The four engines of a commercial aircraft are design so that they each operate independently. Test,
carried out over a long period of time, show that there is a one-in-a hundred chance of in-flight
failure of a single engine.
What is the probability that on a given flight:
a.
No failures occur?
b.
No more than two failures occur?
c.
At least two failures occurs.
d.
Two of the engines run well.
e.
At most one of the engine run well.

BINOMIAL DISTRIBUTION

PROBABILITY OF A STUDENT WILL PASS IN THE

FINAL STATISTICS EXAM = 0.80. IF 6 STUDENTS ARE
SELECTED AT RANDOM, WHAT IS THE
PROBABILITY OF OBTAINING:

a. At most 2 of them are fail in the exam.

b. At least 4 of them will pass in the exam.

Example

If 7 students were selected at random, what is

the probability that:
A. More than 4 students will get score A or C.
B. at least 5 students will get F.
C. less than 3 students will get B or C.
D. all of them will get A or B or F.

Example : 2

a.
b.
c.
d.

Suppose that a family has 5 children. What is the

probability of having:
At least 3 of them are male.
All of them are female.
At most there is one male.
the number of female is more than 4.

Example : 3

a.

b.

c.

Two dice is thrown 5 times, what is the

probability of obtaining:
Total number of both dice is 10 appears at
most twice.
Total number of both dice is 5 appears at
least 4 times.
Total number of both dice is 8 appears
exactly 3 times.

Continuous Probability Distribution

The normal probability distribution has the following

major characteristics :
1. It is bell-shaped and has a single peak.
2. The mean, median and mode are equal.
3. The location of normal distribution is
determined by the and .
4. It falls off smoothly in either direction from the
and the curve gets closer and closer to X
axis but never actually touches it.

NORMAL DISTRIBUTION

1. Xi ~ N ( , )

2. Zi ~ N ( 0, 1 )

What is the difference in meaning between the

first and the second terms.

STANDARD NORMAL VALUE ( Z )

Xi -
Zi = ---------------

Zi MEASURES THE DISTANCE OF ANY

PARTICULAR VALUE OF X FROM THE ,
MEASURED IN UNITS OF THE STANDARD
DEVIATION.
THE SIGN OF Z MIGHT BE POSITIVE, NEGATIVE
OR ZERO. IT DEPENDS ON THE DIFFERENCE
BETWEEN Xi VALUE AND ITS MEAN ( ).

FINDING AREAS UNDER THE NORMAL CURVE

Zi-value
1.65
1.96
1.28
0.50
1.00

PROBABILITY( AREAS )
0.4505
0.4750
0.3997
0.1915
0.3413

Starting point of z value is always from where Zi = 0 up

to the point of another zi value ( the difference between
mean and a selected value of Xi ).

Example: ( Zi value :appendix D, p:720)

a.
b.
c.
d.

e.
f.

The scores of 250 students in mathematics class follow the

normal distribution, with a mean of 69 and standard
deviation of 10. What is the probability or areas under this
normal curve for the scores :
Between 60 and 65.
80
Between 64 and 80.
If the minimum passing grade is 55. how many students fail
in this exam.
What is the minimum scores for 5 % the best.
What is the maximum scores for 37.5% above the average.

Normal Distribution
An electrical firm manufactures light bulbs that have a
length of life that is normally distributed with mean 800
hours and variance of 1600 hours. Find the probability:
a. That a bulb burns between 778 and 834 hours.
b. That a bulb burns 850 hours or 775 hours.
c. That a bulb burns between 720 and 780 hours.
d. The minimum length of life for 12.5% the longest.
e. The minimum length of life for 22% below the average.

MID-EXAM QUESTIONS
1. MEASURES OF CENTRAL TENDENCY:
MEAN, MEDIAN AND MODE AND
COEFFICIENT OF VARIATION.
2. MEASURES OF LOCALITY:
QUARTILE, DECILE AND PERCENTILES.
3. DISPERSION: VARIANCE, STANDARD
DEVIATION AND COEFFICIENT OF
VARIATION FOR UNGROUPED DATA.

MID EXAM QUESTIONS

4. PROBABILITY ( 5 QUESTIONS ):
1.
INDEPENDENT TO CONDITIONAL PROB
2.
BAYES THEOREM
3.
BINOMIAL DISTRIBUTION
4.
NORMAL DISTRIBUTION
5.
BINOMIAL TO NORMAL

MID EXAM QUESTIONS

BINOMIAL DISTRIBUTION ( 2 QUEST)
NORMAL DISTRIBUTION ( 2 QUEST )
YOU MAY SELECT ONLY 4 QUESTIONS.
DURATION = 120 MINUTES
OPEN-BOOK AND NO CHEATING OK?
THERE IS NO MORE MAKE-UP EXAM!
REWARD SYSTEM: CHOCOLATE BAR.
WHEN U FIND ANY DIFFICULTIES PLEASE
DISCUSS WITH ME OR CONTACT TO
0811253224.

The Normal approximation to the Binomial

Binomial distribution is effective if the number of
trials is relatively small. But, generating a binomial
distribution for a large number of trials would be
very time consuming. A more efficient approach is to
apply the normal approximation to the Binomial:
The approach is determined as follows:
=n
Xi n
= n q then, Z =------------ nq

Example

a.

b.

c.

If 10 % of all business executives fill out a given

marketing survey questionnaire, what is the
probability of getting:
At least 15 questionnaires will be back out of 200
distributed to executives?
What is the probability that between 23 and 30
questionnaires will be filled out?
What is the probability that 17 questionnaires or
less will be filled out.

Dont forget to apply the correction factor!

Extra-Bonus
A manufacture produces 10.000 units of product
daily. On the basis of frequent inspections it is
dicovered, 500 unit of product is classified as
defective. What is the probability that:
a.
b.
c.
d.

More than 525 units of product is defective.

At least 9450 units of product is good.
Less than 475 units of product is defective.
The number of good product is between 9450 units
and 9550 units.

The last quiz before mid exam

1. Binomial distribution
2. Normal distribution
3. From Binomial to Normal.
The quiz will be given on Tuesday next
week. ( dont forget to bring your calculator
and statistics tables.

SAMPLING METHODS & ESTIMATION

MODELS

Sampling Methods
REASONS TO SAMPLE
1. TO OBSERVE THE WHOLE POPULATION WOULD BE
TIME CONSUMING AND VERY EXPENSIVE.
2. THE PHYSICAL IMPOSSIBILITY OF CHECKING ALL
ITEMS IN THE POPULATION.
3. THE DESTRUCTIVE NATURE OF SOME TESTS.
4. THE SAMPLE RESULTS ARE ADEQUATE TO INFER
THE CHARACTERISTICS OF POPULATION.

ERROR: sampling error VS non-sampling error

SAMPLING METHODS

1. SIMPLE RANDOM SAMPLE: A SAMPLE SELECTED SO THAT

EACH ITEM IN THE POPULATION HAS THE SAME
PROBABILITY OF BEING INCLUDED.
2. SYSTEMATIC RANDOM SAMPLE: A RANDOM STARTING
POINT k IS SELECTED, AND THEN EVERY kth MEMBER OF
THE POPULATION IS SELECTED.
3. STRATIFIED RANDOM SAMPLE : A POPULATION IS
DIVIDED INTO SUB-GROUPS, CALLED STRATA, AND A
SAMPLE IS RANDOMLY SELECTED FROM EACH STRATUM.
4.CLUSTER SAMPLING: A POPULATION IS DIVIDED INTO
CLUSTERS USING NATURALLY OCCURING GEOGRAPHIC OR
OTHER BOUNDARIES. THEN, CLUSTER ARE RANDOMLY
SELECTED AND A SAMPLE IS COLLECTED BY RANDOMLY
SELECTING FROM EACH CLUSTER.

Sampling Methods

Purposive random sampling

Proportional random sampling
Non-Random sampling
Convenience sampling
Judgment sampling
Snowball sampling

HOW CAN SAMPLES BE USED TO MAKE ESTIMATES?

A SAMPLE IS A TOOL TO INFER SOMETHING ABOUT

POPULATION OR STATISTICS CAN BE USED TO FIND
SOMETHING ABOUT A CHARACTERISTIC OF
POPULATION OR A PARAMETER.

EXAMPLE: X CAN BE USED TO ESTIMATE .

Central limit theorem :

If all samples of a particular size are selected from any
population, the sampling distribution of the sample mean
is approximately a normal distribution. This
approximation improves with larger samples.

SAMPLING DISTRIBUTION OF THE SAMPLE MEAN

SAMPLING DISTRIBUTION OF THE SAMPLE MEAN IS A

PROBABILITY DISTRIBUTION OF ALL POSSIBLE
SAMPLE MEANS OF A GIVEN SAMPLE SIZE ( n ).
THE FOLLOWING EXAMPLE ILLUSTRATES THE
CONTRUCTION OF A SAMPLING DISTRIBUTION OF
SAMPLE MEAN.
LET US ASSUME THAT A POPULATION CONSISTS OF
FOUR ELEMENTS X1 =1, X2 = 2, X3 = 3 AND X4 = 4.
CONSIDER ALL THE SAMPLES OF A GIVEN SIZE n = 2
THAT COULD BE DRAWN FROM THIS POPULATION.
(CONSIDER CAREFULLY WHETHER ONE IS SAMPLING
WITH OR WITHOUT REPLACEMENT).

Sampling with Replacement

All possible
All possible
samples sample means samples sample means
1. 1.1
1.0
9. 3.1
2.0
2. 1.2
1.5
10. 3.2
2.5
3. 1.3
2.0
11. 3.3
3.0
4. 1.4
2.5
12. 3.4
3.5
5. 2.1
1.5
13. 4.1
2.5
6. 2.2
2.0
14. 4.2
3.0
7. 2.3
2.5
15. 4.3
3.5
8 2.4
3.0
16. 4.4
4.0
= ( 1 + 2 + 3 + 4 ) : 4 = 2.5 ( population mean )
x = ( 1 + 1.5 + 2.0 +------------- + 4.0 ) : 16 = 2.5 The mean of all the means
is equal to the population mean or .

Variance and Standard Deviation

2= ( 1 2.5 )2+( 2 2.5 )2 +( 3 2.5 )2 + ( 4 2.5 )2 : 4
= ( 2.25 + 0.25 + 0.25 + 2.25): 4 = 5/4 or 2 = 5/4 = 1.25
= 1.25 = 1.118 ( std deviation of population)
The standard deviation of the sample means is formally referred
as standard error of the mean.
x2=( 1 -2.5 )2 + ( 1.5 2.5 )2 +------------------------- + ( 4 2.5 )2 : 16
= 10 : 16 = 0.625 and x = 0.625 = 0.79 or

1.118
x = ----- = ----------- = 0.79
n
2
Note that when we increase n, the x will decrease.

Without Replacement
All possible samples
Sample means
1.2
1.5
1.3
2.0
1.4
2.5
2.3
2.5
2.4
3.0
3.4
3.5
x = ( 1.5 + 2.0 + ------- + 3.5 ) : 6 = 2.5
x2 = ( 1.5 2.5 )2 + ( 2.0 2.5 )2 +----- + ( 3.5 2.5)2 : 6
= (1 + 0.25 + 0 + 0 + 0.25 + 1) = 2.5 : 6 = 0.417

x = 0.417 = 0.645 or x = --- ( ( N n ) : ( N 1 )

n

Example: Sampling Distribution of The Mean

A population consists of three elements:

X1 = 25, X2= 30 and X3 = 35.
Calculate:
a.
Sampling distribution of the mean if the
sample size = 2 ( with replacement and
without replacement ).
b. Variance and std deviation of the sample
means.

1. Confidence interval for the population mean ( )

when n 30 ( Z value )

Confidence interval for is estimated by:

X a sampling error or X z value SE
Z value depends on confidence level.
SE = standard error of the mean
SE = /n, if std deviation of population is known.
But, if the std deviation of population is unknown,
SE = s/n , s = std deviation of sample.

Sampling Error and Confidence Interval Estimate

Sampling Error : The diffrence between a sample

statistic and its corresponding population parameter.
This error will decrease as the sample size increase.
Confidence interval estimate: A range of values
constructed from sample data so that the population
parameter is likely to within that range occur at
specified probability. The specified probability is
called the level of confidence.

EXAMPLE: Confidence Interval for ( n30 )

The mean and std deviation for the quality grade
point averages of a random samples of 36 college
seniors are calculated to be 2.9 and 0.3
respectively.
a. Find the 92 % confidence interval for the mean of
the entire senior class.
b. How large the sample size is required if we want
to reduce the sampling error not more than 0.05.

Confidence Interval for ( n < 30 )

The weights of 7 similar boxes of cereal are 9.8, 10.2, 10.4,

9.8, 10.0, 10.2 and 9.6 ounces.

Find a 90% confidence interval for he mean of all such boxes

of cereal. ( use t table ).
Step1: calculate the mean and std deviation of sample.
2: use t table to find critical value for 90% C.L ( tv).
3: substitute of all information from step 1 and 2
into the formula:
X - tv ( sd/n) u X + tv ( sd/n )

Level of Confidence
Required level
value of
value of Z table
of confidence
90 %
10 %
1.65
95 %
5%
1.96
99 %
1%
2.58
Example : The prices at which certain type of instant coffee was
being sold on a given day were collected from a random
sample of 45 shops around the country. The mean price was $
1.95 with a standard deviation of $ 0.27. Compute a 80 %
confidence interval for the population mean.

2. A CONFIDENCE INTERVAL FOR A

PROPORTION OF POPULATION ( )
A PROPORTION OF POPULATION ( ) CAN BE
ESTIMATED BY THE FORMULA:
P = X/n
P = THE FRACTION OR PERCENT INDICATING THE PART
OF THE SAMPLE HAVING A PARTICULAR TRAIT OF
INTEREST.
n = SAMPLE SIZE, X IS A PART OF SAMPLE SIZE WHICH
HAVE A PARTICULAR TRAIT OF INTEREST.
CONFIDENCE INTERVAL FOR POPULATION PROPORTION
()

P Zv

P( 1- P ) / n

page:300 no:18

Application

The distribution of household that favor a certain bath soap in

West Java Province (use a 90% confidence level).
Brand
Number of households
Palmolive
1500
Lux
1800
Zest
1500
Beauty
1000
Minty
1200
Other brands
2000
Calculate confidence interval for the proportion of households
that favor LUX bath soap for their families.

4. Confidence interval for the population mean ( ) when the

sample size ( n < 30 ) ttable ( p:722 )

Confidence interval for the is estimated by:

X a sampling error or X t table.SE
SE = S/n ( std deviation of population is unknown).

Example :
The operating life of rechargeable cordless screwdrivers
produced by a firm is assumed to be normally distributed. A
sample of 15 screwdrivers is tested and the mean life is found
to be 8900 hours, with a sample std deviation of 500 hours.
Construct a 90% confidence interval estimate for the
population mean.

Finding t - value
Example : n = 10, confidence level = 90%.

n = 24, confidence level = 95%

Degree of freedom of t- value = df (n-1 )
df
Two-tailed test,
t-value
9
10%
1.833
23
5%
2.069

3. A confidence Interval for The diffrence between Two

Population Means ( 1 - 2 ), n 30

Confidence interval for the diffrence between 1 - 2 is:

( X1 X2 ) Zv x

x = pooled standard deviation.

X1 = the sample mean of X1
X2 = the sample mean of X2
n1 = the sample size of X1
n2 = the sample size of x2

Example
A study was made to estimate the difference in salaries
of college professors in the private and state colleges
of Virginia. A random sample of 100 professor in the
private colleges showed an average of $ 15.000 per
month with a standard deviation of $ 1200. A random
sample of 200 professors in state colleges showed an
average salary of $ 16.000 with a standard deviation
of $ 1400. Find a 90% confidence interval for the
difference between the average salaries of professors
teaching in state and private colleges in Virginia.

5. Estimating the difference between two population

means ( 1- 2 ), n< 30
( X1 - X2) tv SP
A taxi company is trying to decide whether to purchase brand A or brand B tires for its
fleets of taxis. To estimate the difference in the two brands, an experiment is conducted
using 12 tires of brand A and 14 tires of brand B. The results are:
Brand A
Brand B
Average wear-out
22.500 miles
23.600 miles
Std deviation
100 miles
50 miles
Compute 95% confidence interval for the difference between the average quality for the
two brands.

EXAMPLE: POPULATION PROPORTION

A new program for a youth club is planned in a small city. To

determine whether or not the program will get the city
governments support. It is necessary to estimate the proportion
of young people who plan to use the clubs facilities. A survey of
100 randomly selected young people has shown that 22 will use
the facilities if they become available. Construct a 90%
confidence interval for the true proportion.
n = 100, X = 22, P = 22/100 = 0.22 and ( 1 P ) = 0.78
Z value for CL = 90 % = 1.65.

Example:
The performance of stocks between Cement
Industries and Oil industries in 2007 ( in % ).

Cement Industries Oil Industries

Average yields
22
18
Variance
3
2
Sample size
35
40
Calculate interval estimate the difference in average
yields between cement industries and oil industries
( CL = 85 % ).

Quiz 4 : Regression and Correlation

Midterm grades
77
50
71
72
92

Final grades
82
66
78
80
95

a. Construct the regression line by using OLS.

b. Calculate R and R square.
c. When you got score 75 in mid-exam, what is your prediction
score in final-exam?

Hypotheses testing

TESTING HYPOTHESES

( CH: 10 P 116-521)

HYPOTHESIS TESTING: A PROCEDURE BASED ON

SAMPLE EVIDENCE TO DETERMINE WHETHER THE
HYPOTHESIS IS A REASONABLE STATEMENT.
HYPOTHESIS: A STATEMENT ABOUT A POPULATION
DEVELOPED FOR THE PURPOSE OF TESTING.
NULL HYPOTHESIS ( Ho ): SPECIFIES THE VALUE OF
THE PARAMETER TO BE TESTED. e.g: , , ( 1- 2) OR
HYPOTHESES THAT WE FORMULATE WITH THE
HOPE OF REJECTING.
ALTERNATIVE HYPOTHESIS ( Ha ): HYPOTHESES
THAT WE FORMULATE WITH THE HOPE OF
ACCEPTING.

HYPOTHESIS TESTING PROCEDURE

TWO-TAILED TEST: BASICALLY BE USED TO

TEST IF THE QUESTION ASK YOU ABOUT:
THE DIFFERENCE, CHANGE OR EFFECT.
( ALPHA IS DIVIDED BY 2 OR 1/2 )
ONE TAILED TEST : BASICALLY BE USED TO
TEST IF THE QUESTION ASK YOU ABOUT:
GREATER THAN, LESS THAN, INCREASE,
DECREASE, OR MORE THAN. THE CRITICAL
STATISTIC TABLE MAY BE IN THE LEFT OR
IN THE RIGHT TAIL OF THE CURVE.

THE GENERAL PROCEDURE FOR A HYPOTHESIS TEST

START

State the Ho and Ha ( two or one tailed test )

Select the desired CL or alpha

Determine the critical table ( Z, t, F or x2 )

Determine the rejection and acceptance areas of Ho.

Compute the critical ratio by using data sample ( test-statistic )

Compare the critical ratio and table ( accept or reject Ho )
Conclusion

1.Testing the Population Mean ( ), n 30

Example: Mr X, a mouthwash distributor, has stated that the
average cost to process a sales order is $ 13,25. Miss Y, cost
controller, fears that the average cost of processing is more
than $13.25. She is interested in taking action if cost are high,
but she does not care if the actual average cost is below the
assumed value. A random sample of 100 orders had a sample
mean of $13.35, assuming the std deviation is $ 0.50. Conduct
a test at 5 % level of significance and can you conclude that
the average cost of a sales order is more than $13.25?

TEST STATISTIC FOR ( n 30)

Ho=
Ha=
CRITICAL VALUES ( Z TABLE )
ACCEPTANCE & REJECTION AREAS
TEST STATISTIC:
X -
Zh = --------SD/n
. TEST STATISTIC IS FOUND FROM SAMPLE
INFORMATION.
. DECISION: ACCEPT OR REJECT Ho.
. CONCLUSION:

2. TEST STATISTIC FOR

(POPULATION PROPORTION)

TEST STATISTIC:
P
Zh = -------------- ( 1 )/n

TESTING FOR A PROPORTION

POPULATION

RESEARCH AT THE UGM INDICATES THAT 50%

OF THE STUDENTS CHANGE THEIR MAJOR OF
STUDY AFTER THEIR FIRST YEAR IN A
PROGRAM. A RANDOM SAMPLE OF 100
STUDENTS IN THE BUSINESS PROGRAM
REVEALED THAT 48% HAD CHANGED THEIR
MAJOR AREA OF STUDY AFTER THEIR FIRST
YEAR OF THE PROGRAM. Has there been a
significant decrease in the proportion of students who
change their major after the first year in this program?
Test at the 0.05 level of significance.

Example:2
Brand of bath soap Number of Household
Palmolive
1000
Lux
1500
Zest
1000
Beauty
750
Minty
1250
Maya
1000
Other Brands
3500
a. Can you conclude that the proportion of HH that like Lux is
different from 14%.
b. Can you conclude that the proportion of HH that like Lux is
higher than 14 %. ( Alpha = 15 % ).

5. Testing The Population Mean ( ), n < 30

( Use t-table ) p: 785

Test Statistic:
X -
th = -----------SD / n

Example

The distribution of gasoline consumption of 6 sample

Honda Astrea can be reported as follow:
Sample

Range

1
54 km/l
2
53
3
56
4
52
5
50
6
55
The manufacture claimed that the average of gasoline
Consumption was 55 km/l. Can you conclude that this claim is
overestimate? ( use alpha = 5 % ).

TESTING THE DIFFERENCE BETWEEN TWO

POPULATION MEANS

PARAMETRIC STATISTICS:
1. SPSS: PAIRED SAMPLES
2. EVIEWS: UNPAIRED SAMPLES:
EQUALITY OF VARIANCE TEST
3. EXCEL: DATA ANALYSIS
NON PARAMETRIC STATISTICS:
SPSS : RELATED SAMPLES, WILCOXON
SIGN RANK TEST.

3.TEST STATISTIC FOR (1 -2), n 30

(Independent-samples test )

Ho:
Ha:
TEST STATISTIC:
X1 X2
Zh = --------------------------var1/n1 + var2/n2

EXAMPLE

THE PRODUCTION OF ELECTRONIC ITEMS DAILY IN

COMPANY ABC:
MALE WOKERS
FEMALE WORKERS

AVERAGE PROD
VARIANCE
SAMPLE SIZE

450 UNITS
90 UNITS
40 WORKERS

460 UNITS
110 UNITS
50 WORKERS

CAN YOU CONCLUDE THAT THE AVERAGE OF

FEMALE WORKERS PRODUCTIVITY IS HIGHER
THAN MALE WORKERS?
( USE ALPHA = 10 % )

4:Two Sample tests about Proportion ( 1 2 )

A random sample of 100 young women revealed 19 liked
the Heavenly frgrance well enough to purchase it.
Similarly, a sample of 200 older women revealed 62 liked
the fragrance well enough to make a purchase. Can you
conclude that there is a difference in the proportion of
younger and older women who like the Heavenly fragrance?
( alpha = 5%).
Pooled sample proportion Pc = (X1 + X2) / (n1 + n2 )
Test Statistic: Zh = P1 P2 / Sd
Sd = Pc ( 1 Pc ) / n1 + Pc ( 1 Pc ) / n2

6.TEST STATISTIC FOR (1 - 2), n < 30

( USE- t TABLE )

TEST STATISTIC:
X1 - X2
th = ------------------------------------------( n1-1) var1+ (n2-1) var2
1 +1
(n1 + n2 2)
n1 n2

6. Testing (1 - 2 ), n < 30

Dr. Dony, a psychologist, administered IQ tests to

determine if female FEB students were as smart as
male students. The random sample of 15 female
students had a mean score of 131 with std
deviation of 17. The random sample of 13 male
students had a mean of 126 and a std deviation of
14. At 0.01 level of significance:
a. Is there a significant difference in their IQ?
b. Can you conclude that female students IQ is
greater than male students IQ?

7. Dependent Samples test

Sample
1
2
3
4
5
6
7
8
9
10

Productivity of Workers
After Training
Before Training d
( d-d ) ( d d ) 2
235 units
228 units
7
2.4
5.76
210
205
5
0.4
0.16
231
219
12
7.4
54.76
242
240
2
-2.6
6.76
205
198
7
2.4
5.76
230
223
7
2.4
5.76
231
227
4
-0.6
0.36
210
215
-5
-9.6
92.16
225
222
3
-1.6
2.56
249
245
4
-0.6
0.36
Can you conclude that the training program increase the
productivity of the employees? ( alpha = 5% ).

d = 46

SD =

d = 46/10 = 4.6

( d d )2
(n1)
Test statistic:

d
SD / n

Example:2
Sample of customer Score of: New Menu
Old menu
1
36
35
2
48
46
3
50
51
4
76
74
5
55
56
6
60
59
7
71
72
8
66
64
9
62
61
Can you conclude that the new menu is more delicious than
the old one? ( alpha = 5% ).

8: CHY SQUARE TEST

Table: p: 787

Chy- square tests are used in a procedure that

involves the comparison of the differences between
the sample frequencies of the occurrence ( Oij ) and
the hypothetical or theoretical population frequencies
( Eij )
( Expected value ). ( Goodness of Fit ).
It can also be used to test relationship between
variables ( independency test ).
Critical value of x2 depends on the number of rows
and columns

Chy square table

Chy- square table is always one tailed in the

right side of the curve.
Critical table is X2 df ( r -1 )( c - 1 ).
Ho is accepted if test statistics calculated is
less than or equal to its critical table.
Ho is rejected if test statistics calculated is
greater than its critical table.

Chy-Square test

Test statistic:
( Oij Eij )2
Xh2 = -------------Eij
Oij = observed frequencies in ith row & jth column
Eij = expected frequencies in ith row & jth column

EXAMPLE

A GARMENT COMPANY IN CAKUNG RECORDS

THE PERFORMANCE OF ITS LABOUR
PRODUCTIVITY RANDOMLY AS FOLLOWS:
THE LEVEL OF PRODUCTIVITY
LOW MODERATE HIGH
WORK-SHIFT
MORNING (I)
40
45
50
AFTERNOON(II)
60
55
60
NIGHT(III)
40
30
25
Based on the data collected, What is your conclusion?
( ALPHA = 5 % ).

CHY SQUARE: INDEPENDENCY TEST

DISTRIBUTION OF DATA COLLECTED BY UNILEVER

INDONESIA WAS RECORDED AS FOLLOW:

Color of bath soap: Pink White Yellow

Gender: Man
10
10
20

Woman
25
20
5
What is your conclusion? ( UseAlpha = 5 % )

Exercises: check your book p : 661-662-663.

ANALYSIS of VARIANCE

This method can be used to test the difference

among population means ( the number of
specific sample is more than two or k > 2.
The steps of testing the hypothesis.
1. Calculate the variance within samples ( 2w).
2. Calculate variance between or among the
samples (2b).
3. Calculate F table.
4. Calculate F statistic.

9. ANALYSIS OF VARIANCE
FERTILIZER:
SAMPLE
1
2
3
4
5

A
10 kgs
12
13
11
14

12 kgs 11 kgs
11
10
10
9
10
10
12
10

D
9 kgs
10
8
8
10

Can you conclude that the average productivity of those

fertilizers are significantly different? ( alpha = 5% ).