Chapter 9

The MI Procedure

Chapter Table of Contents

OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
GETTING STARTED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
SYNTAX . . . . . . . . .
PROC MI Statement . .
BY Statement . . . . . .
EM Statement . . . . . .
FREQ Statement . . . .
MCMC Statement . . . .
MONOTONE Statement
TRANSFORM Statement
VAR Statement . . . . .

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

137
138
141
141
142
143
149
150
151

DETAILS . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Descriptive Statistics . . . . . . . . . . . . . . . . . . . . .
EM Algorithm for Data with Missing Values . . . . . . . . .
Statistical Assumptions for Multiple Imputation . . . . . . .
Missing Data Patterns . . . . . . . . . . . . . . . . . . . . .
Imputation Mechanisms . . . . . . . . . . . . . . . . . . .
Regression Method for Monotone Missing Data . . . . . . .
Propensity Score Method for Monotone Missing Data . . . .
MCMC Method for Arbitrary Missing Data . . . . . . . . .
Producing Monotone Missingness with the MCMC Method .
MCMC Method Specifications . . . . . . . . . . . . . . . .
Convergence in MCMC . . . . . . . . . . . . . . . . . . . .
Input Data Sets . . . . . . . . . . . . . . . . . . . . . . . .
Output Data Sets . . . . . . . . . . . . . . . . . . . . . . .
Combining Inferences from Multiply Imputed Data Sets . .
Multiple Imputation Efficiency . . . . . . . . . . . . . . . .
Imputers Model Versus Analysts Model . . . . . . . . . .
Parameter Simulation Versus Multiple Imputation . . . . . .
ODS Table Names . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

152
152
153
154
155
156
157
158
159
164
166
167
170
171
173
174
174
175
176

EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Example 9.1 EM Algorithm for MLE . . . . . . . . . . . . . . . . . . . . . 177

130 

Chapter 9. The MI Procedure

Example 9.2 Propensity Score Method . . . . . . . . . . . .
Example 9.3 Regression Method . . . . . . . . . . . . . . .
Example 9.4 MCMC Method . . . . . . . . . . . . . . . . .
Example 9.5 Producing Monotone Missingness with MCMC
Example 9.6 Checking Convergence in MCMC . . . . . . .
Example 9.7 Transformation to Normality . . . . . . . . . .
Example 9.8 Saving and Using Parameters for MCMC . . .

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

181
184
185
188
191
194
198

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199

SAS OnlineDoc: Version 8

Chapter 9

The MI Procedure
Overview
The experimental MI procedure performs multiple imputation of missing data. Missing values are an issue in a substantial number of statistical analyses. Most SAS
statistical procedures exclude observations with any missing variable values from
the analysis. These observations are called incomplete cases. While analyzing only
complete cases has its simplicity, the information contained in the incomplete cases
is lost. This approach also ignores possible systematic differences between the complete cases and the incomplete cases, and the resulting inference may not be applicable to the population of all cases, especially with a smaller number of complete
cases.
Some SAS procedures use all the available cases in an analysis, that is, cases with
available information. For example, the CORR procedure estimates a variable mean
by using all cases with nonmissing values for this variable, ignoring the possible
missing values in other variables. PROC CORR also estimates a correlation by using
all cases with nonmissing values for this pair of variables. This makes better use of
the available data, but the resulting correlation matrix may not be positive definite.
Another strategy for handling missing data is simple imputation, which substitutes a
value for each missing value. Standard statistical procedures for complete data analysis can then be used with the filled-in data set. For example, each missing value
can be imputed with the variable mean of the complete cases, or it can be imputed
with the mean conditional on observed values of other variables. This approach treats
missing values as if they were known in the complete-data analysis. However, single imputation does not reflect the uncertainty about the predictions of the unknown
missing values, and the resulting estimated variances of the parameter estimates will
be biased toward zero (Rubin 1987, p. 13).
Instead of filling in a single value for each missing value, multiple imputation (Rubin
1976; 1987) replaces each missing value with a set of plausible values that represent
the uncertainty about the right value to impute. The multiply imputed data sets are
then analyzed by using standard procedures for complete data and combining the
results from these analyses. No matter which complete-data analysis is used, the
process of combining results from different data sets is essentially the same.
Multiple imputation does not attempt to estimate each missing value through simulated values but rather to represent a random sample of the missing values. This
process results in valid statistical inferences that properly reflect the uncertainty due
to missing values; for example, confidence intervals with the correct probability coverage.

132 

Chapter 9. The MI Procedure

Multiple imputation inference involves three distinct phases:
1. The missing data are filled in m times to generate m complete data sets.
2. The m complete data sets are analyzed using standard statistical analyses.
3. The results from the m complete data sets are combined to produce inferential
results.
The new MI procedure creates multiply imputed data sets for incomplete multivariate
data. It uses methods that incorporate appropriate variability across the m imputations. The method of choice depends on the patterns of missingness. A data set with
variables Y1 , Y2 , ..., Yp (in that order) is said to have a monotone missing pattern
when the event that a variable Yj is missing for a particular individual implies that all
subsequent variables Yk , k > j , are missing for that individual.
For data sets with monotone missing patterns, either a parametric regression method
(Rubin 1987) that assumes multivariate normality or a nonparametric method that
uses propensity scores (Rubin 1987; Lavori, Dawson, and Shera 1995) is appropriate. For data sets with arbitrary missing patterns, a Markov Chain Monte Carlo
(MCMC) method (Schafer 1997) that assumes multivariate normality is used to impute all missing values or just enough missing values to make the imputed data sets
have monotone missing patterns.
Once the m complete data sets are analyzed using standard SAS procedures, the new
MIANALYZE procedure can be used to generate valid statistical inferences about
these parameters by combining results from the m analyses. These two procedures
are available in experimental form in Release 8.2 of the SAS System.
Often, as few as three to five imputations are adequate in multiple imputation (Rubin
1996, p. 480). The relative efficiency of the small m imputation estimator is high for
cases with little missing information (Rubin 1987, p. 114). Also see the Multiple
Imputation Efficiency section on page 174.
Multiple imputation inference assumes that the model (variables) you used to analyze
the multiply imputed data (the analysts model) is the same as the model used to impute missing values in multiple imputation (the imputers model). But in practice, the
two models may not be the same. The consequence for different scenarios (Schafer
1997, pp. 139143) is discussed in the Imputers Model Versus Analysts Model
section on page 174.
In addition to the multiple imputation method, a simulation-based method of parameter simulation can also be used to analyze the data for many incomplete-data
problems. Although the MI procedure does not offer a simulation-based method of
parameter simulation, the choice between the two methods (Schafer 1997, pp. 8990,
135136) is examined in the Parameter Simulation Versus Multiple Imputation section on page 175.

SAS OnlineDoc: Version 8

Getting Started

133

Getting Started
Consider the following Fitness data set that has been altered to contain an arbitrary
pattern of missingness:
*----------------- Data on Physical Fitness -----------------*
| These measurements were made on men involved in a physical |
| fitness course at N.C. State University.
|
| Only selected variables of
|
| Oxygen (oxygen intake, ml per kg body weight per minute), |
| Runtime (time to run 1.5 miles in minutes), and
|
| RunPulse (heart rate while running) are used.
|
| Certain values were changed to missing for the analysis.
|
*------------------------------------------------------------*;
data FitMiss;
input Oxygen RunTime RunPulse @@;
datalines;
44.609 11.37 178
45.313 10.07
54.297
8.65 156
59.571
.
49.874
9.22
.
44.811 11.63
.
11.95 176
. 10.85
39.442 13.08 174
60.055
8.63
50.541
.
.
37.388 14.03
44.754 11.12 176
47.273
.
51.855 10.33 166
49.156
8.95
40.836 10.95 168
46.672 10.00
46.774 10.25
.
50.388 10.08
39.407 12.63 174
46.080 11.17
45.441
9.63 164
.
8.92
45.118 11.08
.
39.203 12.88
45.790 10.47 186
50.545
9.93
48.673
9.40 186
47.920 11.50
47.467 10.50 170
;

185
.
176
.
170
186
.
180
.
168
156
.
168
148
170

Suppose that the data are multivariate normally distributed and the missing data are
missing at random (MAR). That is, the probability that an observation is missing
can depend on the observed variable values of the individual, but not on the missing variable values of the individual. See the Statistical Assumptions for Multiple
Imputation section on page 154 for a detailed description of the MAR assumption.
The following statements invoke the MI procedure and impute missing values for the
FitMiss data set.
proc mi data=FitMiss seed=37851 mu0=50 10 180 out=outmi;
var Oxygen RunTime RunPulse;
run;

SAS OnlineDoc: Version 8

134 

Chapter 9. The MI Procedure

The MI Procedure
Model Information
Data Set
Method
Multiple Imputation Chain
Initial Estimates for MCMC
Start
Prior
Number of Imputations
Number of Burn-in Iterations
Number of Iterations
Seed for random number generator

Figure 9.1.

WORK.FITMISS
MCMC
Single Chain
EM Posterior Mode
Starting Value
Jeffreys
5
200
100
37851

Model Information

The Model Information table describes the method used in the multiple imputation
process. By default, the procedure uses the Markov Chain Monte Carlo (MCMC)
method with a single chain to create five imputations. The posterior mode, the highest
observed-data posterior density, with a noninformative prior, is computed from the
EM algorithm and is used as the starting value for the chain.
The MI procedure takes 200 burn-in iterations before the first imputation and 100
iterations between imputations. In a Markov chain, the information in the current
iteration has influence on the state of the next iteration. The burn-in iterations are
iterations in the beginning of each chain that are used both to eliminate the series of
dependence on the starting value of the chain and to achieve the stationary distribution. The between-imputation iterations in a single chain are used to eliminate the
series of dependence between the two imputations.
The MI Procedure
Missing Data Patterns

Group
1
2
3
4
5

Oxygen

Run
Time

Run
Pulse

X
X
X
.
.

X
X
.
X
X

X
.
.
X
.

Freq

Percent

21
4
3
1
2

67.74
12.90
9.68
3.23
6.45

Missing Data Patterns

-----------------Group Means---------------Oxygen
RunTime
RunPulse

Group
1
2
3
4
5

Figure 9.2.

46.353810
47.109500
52.461667
.
.

Missing Data Patterns

SAS OnlineDoc: Version 8

10.809524
10.137500
.
11.950000
9.885000

171.666667
.
.
176.000000
.

Getting Started

135

The Missing Data Patterns table lists distinct missing data patterns with corresponding frequencies and percents. Here, an X means that the variable is observed
in the corresponding group and a . means that the variable is missing. The table also
displays group-specific variable means. The MI procedure sorts the data into groups
based on whether an individuals value is observed or missing for each variable to be
analyzed. For a detailed description of missing data patterns, see the Missing Data
Patterns section on page 155.
The MI Procedure
Multiple Imputation Variance Information

Variable
Oxygen
RunTime
RunPulse

-----------------Variance----------------Between
Within
Total
0.045321
0.005853
0.611864

0.937239
0.072217
3.247163

0.991624
0.079241
3.981400

DF
26.113
24.45
19.227

Multiple Imputation Variance Information

Figure 9.3.

Variable

Relative
Increase
in Variance

Fraction
Missing
Information

Oxygen
RunTime
RunPulse

0.058027
0.097265
0.226116

0.056263
0.092202
0.197941

Variance Information

After the completion of m imputations, the Multiple Imputation Variance Information table displays the between-imputation variance, within-imputation variance,
and total variance for combining complete-data inferences. It also displays the degrees of freedom for the total variance. The relative increase in variance due to
missing values and the fraction of missing information for each variable are also
displayed. A detailed description of these statistics is provided in the Combining
Inferences from Multiply Imputed Data Sets section on page 173.
The following Multiple Imputation Parameter Estimates table displays the estimated mean and standard error of the mean for each variable. The inferences are
based on the t distribution. The table also displays a 95% confidence interval for the
mean and a t statistic with the associated p-value for the hypothesis that the population
mean is equal to the value specified with the MU0= option. A detailed description
of these statistics is provided in the Combining Inferences from Multiply Imputed
Data Sets section on page 173.

SAS OnlineDoc: Version 8

136 

Chapter 9. The MI Procedure

The MI Procedure
Multiple Imputation Parameter Estimates
Variable

Mean

Std Error

Oxygen
RunTime
RunPulse

47.126919
10.546494
171.621676

0.995803
0.281498
1.995344

95% Confidence Limits

45.0804
9.9661
167.4487

49.1734
11.1269
175.7946

DF
26.113
24.45
19.227

Multiple Imputation Parameter Estimates

Variable

Minimum

Maximum

Mu0

t for H0:
Mean=Mu0

Pr > |t|

Oxygen
RunTime
RunPulse

46.849494
10.464123
170.623678

47.318758
10.669193
172.680679

50.000000
10.000000
180.000000

-2.89
1.94
-4.20

0.0077
0.0638
0.0005

Figure 9.4.

Parameter Estimates

In addition to the output tables, the procedure also creates a data set with imputed
values. The imputed data sets are stored in the outmi data set, with the index variable
Imputation indicating the imputation numbers. The data set can now be analyzed
using standard statistical procedures with Imputation as a BY variable.
The following statements list the first ten observations of data set outmi.
proc print data=outmi (obs=10);
title First 10 Observations of the Imputed Data Set;
run;

First 10 Observations of the Imputed Data Set

Obs
1
2
3
4
5
6
7
8
9
10

Figure 9.5.

_Imputation_

Oxygen

RunTime

Run
Pulse

1
1
1
1
1
1
1
1
1
1

44.6090
45.3130
54.2970
59.5710
49.8740
44.8110
46.0264
42.3040
39.4420
60.0550

11.3700
10.0700
8.6500
6.1569
9.2200
11.6300
11.9500
10.8500
13.0800
8.6300

178.000
185.000
156.000
138.583
164.163
176.000
176.000
182.486
174.000
170.000

Imputed Data Set

The table shows that the precision of the imputed values differs from the precision of
the observed values. You can use the ROUND= option to make the imputed values
consistent with the observed values.

SAS OnlineDoc: Version 8

Syntax

137

Syntax
The following statements are available in PROC MI.

PROC MI < options > ;

BY variables ;
EM < options > ;
FREQ variable ;
MCMC < options > ;
MONOTONE < options > ;
TRANSFORM transform ( variables < / options >)
< : : : transform ( variables < / options >) > ;
VAR variables ;

The BY statement specifies groups in which separate multiple imputation analyses

are performed.
The EM statement uses the EM algorithm to compute the maximum likelihood estimate (MLE) of the data with missing values, assuming a multivariate normal distribution for the data.
The FREQ statement specifies the variable that represents the frequency of occurrence for other values in the observation.
The MCMC statement uses a Markov chain Monte Carlo method to impute values for
a data set with an arbitrary missing pattern. The MONOTONE statement uses either a
parametric regression method or a nonparametric method based on propensity scores
to impute values for a data set with a monotone missing pattern. Note that you can use
either an MCMC statement or a MONOTONE statement, but not both. When neither
of these two statements is specified, the MCMC method with its default options is
used.
The TRANSFORM statement lists the variables to be transformed before the imputation process. The imputed values of these transformed variables will be reversetransformed to the original forms before the imputation.
The VAR statement lists the numeric variables to be analyzed. If you omit the VAR
statement, all numeric variables not listed in other statements are used.
The PROC MI statement is the only required statement for the MI procedure. The
rest of this section provides detailed syntax information for each of these statements,
beginning with the PROC MI statement. The remaining statements are in alphabetical
order.

SAS OnlineDoc: Version 8

138 

Chapter 9. The MI Procedure

PROC MI Statement
PROC MI < options > ;
The following table summarizes the options available in the PROC MI statement.
Table 9.1.

Summary of PROC MI Options

Tasks

Options

Specify data sets

input data set
output data set with imputed values

DATA=
OUT=

Specify imputation details

number of imputations
seed to begin random number generator
units to round imputed variable values
maximum values for imputed variable values
minimum values for imputed variable values
singularity tolerance

NIMPUTE=
SEED=
ROUND=
MAXIMUM=
MINIMUM=
SINGULAR=

Specify statistical analysis

level for the confidence interval, (1
means under the null hypothesis

ALPHA=
MU0=

Control printed output

suppress all displayed output
displays univariate statistics and correlations

NOPRINT
SIMPLE

The following options can be used in the PROC MI statement (in alphabetical order):
ALPHA=

specifies that confidence limits be constructed for the mean estimates with confidence
level 100(1 )%, where 0 <  < 1. The default is ALPHA=0.05.
DATA=SAS-data-set

names the SAS data set to be analyzed by PROC MI. By default, the procedure uses
the most recently created SAS data set.
MAXIMUM=numbers

specifies maximum values for imputed variables. When an intended imputed value
is greater than the maximum, PROC MI redraws another value for imputation. If
only one number is specified, that number is used for all variables. If more than one
number is specified, you must use a VAR statement, and the specified numbers must
correspond to variables in the VAR statement. A missing value indicates no restriction on the maximum for the corresponding variable. The default is MAXIMUM=. ,
no restriction on the maximum.

SAS OnlineDoc: Version 8

PROC MI Statement

139

The MAXIMUM= option is related to the MINIMUM= and ROUND= options,

which are used to make the imputed values more consistent with the observed variable
values. These options are not applicable if you specify the METHOD=PROPENSITY
option in the MONOTONE statement.
When specifying a maximum for the first variable only, you must also specify a missing value after the maximum. Otherwise, the maximum is used for all variables.
For example, the MAXIMUM= 100 . option sets a maximum of 100 for the first
analysis variable only and no maximum for the remaining variables. The MAXIMUM= . 100 option sets a maximum of 100 for the second analysis variable only
and no maximum for the other variables.
MINIMUM=numbers

specifies the minimum values for imputed variables. When an intended imputed value
is less than the minimum, PROC MI redraws another value for imputation. If only one
number is specified, that number is used for all variables. If more than one number is
specified, you must use a VAR statement, and the specified numbers must correspond
to variables in the VAR statement. A missing value indicates no restriction on the
minimum for the corresponding variable. The default is MINIMUM=. , no restriction
on the minimum.
MU0=numbers
THETA0=numbers

specifies the parameter values 0 under the null hypothesis  = 0 for the population
means corresponding to the analysis variables. Each hypothesis is tested with a t test.
If only one number is specified, that number is used for all variables. If more than
one number is specified, you must use a VAR statement, and the specified numbers
must correspond to variables in the VAR statement. The default is MU0=0.
If a variable is transformed as specified in a TRANSFORM statement, then the same
transformation for that variable is also applied to its corresponding specified MU0=
value in the t test. If the parameter values 0 for a transformed variable is not specified, then 0 = 0 is used for that transformed variable.

NIMPUTE=number

specifies the number of imputations. The default is NIMPUTE=5. You can specify
NIMPUTE=0 to skip the imputation. In this case, only tables of model information,
missing data patterns, descriptive statistics (SIMPLE option), and MLE from the EM
algorithm (EM statement) are displayed.
NOPRINT

suppresses the display of all output. Note that this option temporarily disables the
Output Delivery System (ODS). For more information, refer to the chapter Using
the Output Delivery System in the SAS/STAT Users Guide, Version 8.
OUT=SAS-data-set

creates an output SAS data set containing imputation results. The data set includes
an index variable, Imputation , to identify the imputation number. For each imputation, the data set contains all variables in the input data set with missing values
replaced by the imputed values. See the Output Data Sets section on page 171 for
a description of this data set.

SAS OnlineDoc: Version 8

140 

Chapter 9. The MI Procedure

If you want to create a permanent SAS data set, you must specify a two-level name.
For more information on permanent SAS data sets, refer to the section SAS Files
in SAS Language Reference: Concepts, Version 8.

ROUND=numbers

specifies the units to round variables in the imputation. If only one number is specified, that number is used for all variables. If more than one number is specified, you
must use a VAR statement, and the specified numbers must correspond to variables
in the VAR statement. The default number is a missing value, which indicates no
rounding for imputed variables.
When specifying a roundoff unit for the first variable only, you must also specify a
missing value after the roundoff unit. Otherwise, the roundoff unit is used for all
variables. For example, the option ROUND= 10 . sets a roundoff unit of 10 for the
first analysis variable only and no rounding for the remaining variables. The option
ROUND= . 10 sets a roundoff unit of 10 for the second analysis variable only and
no rounding for other variables.
You can use the ROUND= option to set the precision of imputed values. For example, with a roundoff unit of 0.001, each value is rounded to the nearest multiple of
0.001. That is, each value has three significant digits after the decimal point. See
Example 9.3 for a usage of this option.
SEED=number

specifies a positive integer. PROC MI uses the value of the SEED= option to start
the pseudo-random number generator. The default is a value generated from reading
the time of day from the computers clock. However, in order to duplicate the results
under identical situations, you must control the value of the seed explicitly rather than
rely on the clock reading.
The seed information is displayed in the Model Information table so that the results
can be reproduced by specifying this seed with the SEED= option. You need to
specify the same seed number in the future to reproduce the results.
SIMPLE

displays simple descriptive univariate statistics and pairwise correlations from available cases. For a detailed description of these statistics, see the Descriptive
Statistics section on page 152.
SINGULAR=p

specifies the criterion for determining the singularity of a covariance matrix, where
0 < p < 1. The default is SINGULAR=1E 8.
Suppose that S is a covariance matrix and v is the number of variables in S. Based on
the spectral decomposition S =  0 , where  is a diagonal matrix of eigenvalues
j , j = 1; : : :, v, where i  j when i < j , and is a matrix with the corresponding orthonormal eigenvectors of S as columns, S is
singular when
Pconsidered
v


an eigenvalue j is less than p, where the average  = k=1 k =v .

SAS OnlineDoc: Version 8

EM Statement

141

BY Statement
BY variables ;
You can specify a BY statement with PROC MI to obtain separate analyses on observations in groups defined by the BY variables. When a BY statement appears, the
procedure expects the input data set to be sorted in order of the BY variables.
If your input data set is not sorted in ascending order, use one of the following alternatives:




Sort the data using the SORT procedure with a similar BY statement.
Specify the BY statement option NOTSORTED or DESCENDING in the BY
statement for the MI procedure. The NOTSORTED option does not mean that
the data are unsorted but rather that the data are arranged in groups (according to values of the BY variables) and that these groups are not necessarily in
alphabetical or increasing numeric order.
Create an index on the BY variables using the DATASETS procedure.

For more information on the BY statement, refer to the discussion in SAS Language
Reference: Concepts, Version 8. For more information on the DATASETS procedure,
refer to the discussion in the SAS Procedures Guide, Version 8.

EM Statement
EM < options > ;
The expectation-maximization (EM) algorithm is a technique for maximum likelihood estimation in parametric models for incomplete data. The EM statement uses
the EM algorithm to compute the MLE for (; ), the means and covariance matrix, of a multivariate normal distribution from the input data set with missing values.
PROC MI uses the means and standard deviations from available cases as the initial
estimates for the EM algorithm. The correlations are set to zero.
You can also use the EM statement with the NIMPUTE=0 option in the PROC statement to compute the EM estimates without multiple imputation, as shown in Example 9.1 in the Examples section on page 177.

SAS OnlineDoc: Version 8

142 

Chapter 9. The MI Procedure

The following five options are available with the EM statement.
CONVERGE=p

sets the convergence criterion. The value must be between 0 and 1. The iterations are
considered to have converged when the maximum change in the parameter estimates
between iteration steps is less than the value specified. The change is a relative change
if the parameter is greater than 0.01 in absolute value; otherwise, it is an absolute
change. By default, CONVERGE=1E-4.
ITPRINT

prints the iteration history in the EM algorithm.

MAXITER=number

specifies the maximum number of iterations used in the EM algorithm. The default
is MAXITER=200.
OUTEM=SAS-data-set

creates an output SAS data set of TYPE=COV containing the MLE of the parameter
vector (; ). These estimates are computed with the EM algorithm. See the Output
Data Sets section on page 171 for a description of this output data set.
OUTITER < ( options ) > =SAS-data-set

creates an output SAS data set of TYPE=COV containing parameters for each iteration. The data set includes a variable named Iteration to identify the iteration
number.
The parameters in the output data set depend on the options specified. You can specify
the MEAN and COV options to output the mean and covariance parameters. When
no options are specified, the output data set contains the mean parameters for each
iteration. See the Output Data Sets section on page 171 for a description of this
data set.

FREQ Statement
FREQ variable ;
If one variable in your input data set represents the frequency of occurrence for other
values in the observation, specify the variable name in a FREQ statement. PROC MI
then treats the data set as if each observation appears n times, where n is the value of
the FREQ variable for the observation. If the value of the FREQ variable is less than
one, the observation is not used in the analysis. Only the integer portion of the value
is used. The total number of observations is considered to be equal to the sum of the
FREQ variable when PROC MI calculates significance probabilities.

SAS OnlineDoc: Version 8

MCMC Statement

143

MCMC Statement
MCMC < options > ;
The MCMC statement specifies the details of the MCMC method for imputation. The
following table summarizes the options available for the MCMC statement.
Table 9.2.

Summary of Options in MCMC

Tasks

Options

Specify data sets

input parameter estimates for imputations
output parameter estimates used in imputations
output parameter estimates used in iterations

INEST=
OUTEST=
OUTITER=

Specify imputation details

monotone/full imputation
single/multiple chain
number of burn-in iterations for each chain
number of iterations between imputations in a chain
initial parameter estimates for MCMC
prior parameter information
starting parameters

IMPUTE=
CHAIN=
NBITER=
NITER=
INITIAL=
PRIOR=
START=

Specify output graphics

displays time-series plots
displays autocorrelation plots
graphics catalog name for saving graphics output

TIMEPLOT=
ACFPLOT=
GOUT=

Control printed output

displays worst linear function
displays initial parameter values for MCMC

WLF
DISPLAYINIT

The following are the options available for the MCMC statement (in alphabetical
order):
ACFPLOT < ( options < / display-options > ) >

displays the autocorrelation function plots of parameters from iterations.

The available options are:
COV < ( < variables > < variable1*variable2

> < : : : variable1*variable2 > ) >

displays plots of variances for variables in the list and covariances for pairs
of variables in the list. When the option COV is specified without variables,
variances for all variables and covariances for all pairs of variables are used.

SAS OnlineDoc: Version 8

144 

Chapter 9. The MI Procedure

MEAN < ( variables ) >

displays plots of means for variables in the list. When the option MEAN is
specified without variables, all variables are used.
WLF

displays the plot for the worst linear function.

When the ACFPLOT is specified without the preceding options, the procedure displays plots of means for all variables that are used.
The display-options provide additional information for the autocorrelation function
plots. The available display-options are:
CCONF=color

specifies the color of the displayed confidence limits.

CCONF=BLACK.

The default is

CFRAME=color

specifies the color for filling the area enclosed by the axes and the frame. By
default, this area is not filled.
CNEEDLES=color

specifies the color of the vertical line segments (needles) that connect autocorrelations to the reference line. The default is CNEEDLES=BLACK.
CREF=color

specifies the color of the displayed reference line.

CREF=BLACK.

The default is

CSYMBOL=color

specifies the color of the displayed data points.

BOL=BLACK.

The default is CSYM-

HSYMBOL=number

specifies the height for data points in percentage screen units. The default is
HSYMBOL=1.
LCONF=linetype

specifies the line type for the displayed confidence limits. The default is
LREF=1, a solid line.
LOG

requests that the logarithmic transformations of parameters be used to compute

the autocorrelations. Its generally used for the variances of variables. When
a parameter has values less than or equal to zero, the corresponding plot is not
created.
LREF=linetype

specifies the line type for the displayed reference line. The default is LREF=3,
a dashed line.
NLAG=number

specifies the maximum lag of the series. The default is NLAG=20. The autocorrelations at each lag are displayed in the graph.

SAS OnlineDoc: Version 8

MCMC Statement

145

SYMBOL=value

specifies the symbol for data points in percentage screen units. The default is
SYMBOL=STAR.
TITLE=string

specifies the title to be displayed in the autocorrelation function plots. The

default is TITLE=Autocorrelation Plot.
WCONF=number

specifies the width for the displayed confidence limits in percentage screen
units. If you specify the WCONF=0 option, the confidence limits are not displayed. The default is WCONF=1.
WNEEDLES=number

specifies the width for the displayed needles that connect autocorrelations to
the reference line in percentage screen units. If you specify the WNEEDLES=0
option, the needles are not displayed. The default is WNEEDLES=1.
WREF=number

specifies the width for the displayed reference line in percentage screen units.
If you specify the WREF=0 option, the reference line is not displayed. The
default is WREF=1.
For example, the statement
acfplot( mean( y1) cov(y1) /log);

requests autocorrelation function plots for the means and variances of the variable y1, respectively. Logarithmic transformations of both the means and variances are used in the plots. For a detailed description of the autocorrelation
function plot, see the Autocorrelation Function Plot section on page 169; refer also to Schafer (1997, pp. 120-126) and the SAS/ETS Users Guide, Version
8.

CHAIN=SINGLE | MULTIPLE

specifies whether a single chain is used for all imputations or a separate chain is used
for each imputation. The default is CHAIN=SINGLE.
DISPLAYINIT

displays initial parameter values in the MCMC process for each imputation.
GOUT=graphics-catalog

specifies the graphics catalog for saving graphics output from PROC MI. The default is WORK.GSEG. For more information, refer to the chapter The GREPLAY
Procedure in SAS/GRAPH Software: Reference, Version 8.
IMPUTE=FULL | MONOTONE

specifies whether a full-data imputation is used for all missing values or a monotonedata imputation is used for a subset of missing values to make the imputed data sets
have a monotone missing pattern. The default is IMPUTE=FULL. When
IMPUTE=MONOTONE is specified, the order in the VAR statement is used to complete the monotone pattern.

SAS OnlineDoc: Version 8

146 

Chapter 9. The MI Procedure

INEST=SAS-data-set

names a SAS data set of TYPE=EST containing parameter estimates for imputations.
These estimates are used to impute values for observations in the DATA= data set.
A detailed description of the data set is provided in the Input Data Sets section on
page 170.
INITIAL=EM < ( options ) >
INITIAL=INPUT=SAS-data-set

specifies the initial mean and covariance estimates for the MCMC process. The default is INITIAL=EM.
You can specify INITIAL=INPUT=SAS-data-set to read the initial estimates of the
mean and covariance matrix for each imputation from a SAS data set. See the Input
Data Sets section on page 170 for a description of this data set.
With INITIAL=EM, PROC MI derives parameter estimates for a posterior mode,
the highest observed-data posterior density, from the EM algorithm. The MLE from
EM is used to start the EM algorithm for the posterior mode, and the resulting EM
estimates are used to begin the MCMC process.
The following four options are available with INITIAL=EM.
BOOTSTRAP < =number >

requests bootstrap resampling, which uses a simple random sample with replacement from the input data set for the initial estimate. You can explicitly
specify the number of observations in the random sample. Alternatively, you
can implicitly specify the number of observations in the random sample by
specifying the proportion p; 0 < p <= 1, to request [np] observations in the
random sample, where n is the number of observations in the data set and [np]
is the integer part of np. This produces an overdispersed initial estimate that
provides different starting values for the MCMC process. If you specify the
BOOTSTRAP option without the number, p=0.75 is used by default.
CONVERGE=p

sets the convergence criterion. The value must be between 0 and 1. The iterations are considered to have converged when the maximum change in the
parameter estimates between iteration steps is less than the value specified. The
change is a relative change if the parameter is greater than 0.01 in absolute
value; otherwise, it is an absolute change. By default, CONVERGE=1E-4.
ITPRINT

prints the iteration history in the EM algorithm for the posterior mode.
MAXITER=number

specifies the maximum number of iterations used in the EM algorithm. The

default is MAXITER=200.

SAS OnlineDoc: Version 8

MCMC Statement

147

NBITER=number

specifies the number of burn-in iterations before the first imputation in each chain.
The default is NBITER=200.
NITER=number

specifies the number of iterations between imputations in a single chain. The default
is NITER=100.
OUTEST=SAS-data-set

creates an output SAS data set of TYPE=EST. The data set contains parameter
estimates used in each imputation. The data set also includes a variable named
Imputation to identify the imputation number. See the Output Data Sets section
on page 171 for a description of this data set.
OUTITER < ( options ) > =SAS-data-set

creates an output SAS data set of TYPE=COV containing parameters used in the imputation step for each iteration. The data set includes variables named Imputation
and Iteration to identify the imputation number and iteration number.
The parameters in the output data set depend on the options specified. You can specify options MEAN, STD, COV, LR, LR POST, and WLF to output parameters of
means, standard deviations, covariances, -2 log LR statistic, -2 log LR statistic of the
posterior mode, and the worst linear function. When no options are specified, the
output data set contains the mean parameters used in the imputation step for each
iteration. See the Output Data Sets section on page 171 for a description of this
data set.
PRIOR=name

specifies the prior information for the means and covariances. Valid values for name
are as follows:
JEFFREYS

specifies a noninformative prior.

RIDGE=number

specifies a ridge prior.

INPUT=SAS-data-set

specifies a data set containing prior information.

For a detailed description of the prior information, see the Bayesian Estimation of
the Mean Vector and Covariance Matrix section on page 161 and the Posterior
Step section on page 162. If you do not specify the PRIOR= option, the default is
PRIOR=JEFFREYS.
The PRIOR=INPUT= option specifies a TYPE=COV data set from which the prior
information of the mean vector and the covariance matrix is read. See the Input Data
Sets section on page 170 for a description of this data set.
START=VALUE | DIST

specifies that the initial parameter estimates are used as either the starting value
(START=VALUE) or as the starting distribution (START=DIST) in the first imputation step of each chain. The default is START=VALUE.

SAS OnlineDoc: Version 8

148 

Chapter 9. The MI Procedure

TIMEPLOT < ( options < / display-options > ) >

displays the time-series plots of parameters from iterations.

The available options are:
COV < ( < variables > < variable1*variable2

> < : : : variable1*variable2 > ) >

displays plots of variances for variables in the list and covariances for pairs
of variables in the list. When the option COV is specified without variables,
variances for all variables and covariances for all pairs of variables are used.
MEAN < ( variables ) >

displays plots of means for variables in the list. When the option MEAN is
specified without variables, all variables are used.
WLF

displays the plot for the worst linear function.

When the TIMEPLOT is specified without the preceding options, the procedure displays plots of means for all variables are used.
The display-options provide additional information for the time-series plots. The
available display-options are:
CFRAME=color

specifies the color for filling the area enclosed by the axes and the frame. By
default, this area is not filled.
CSYMBOL=color

specifies the color of the data points to be displayed in the time-series plots.
The default is CSYMBOL=BLACK.
HSYMBOL=number

specifies the height for data points in percentage screen units. The default is
HSYMBOL=1.
LOG

requests that the logarithmic transformations of parameters be used. Its generally used for the variances of variables. When a parameter value is less than or
equal to zero, the value is not displayed in the corresponding plot.
SYMBOL=value

specifies the symbol for data points in percentage screen units. The default is
SYMBOL=PLUS.
TITLE=string

specifies the title to be displayed in the time-series plots.

TITLE=Time-series Plot for Iterations.

The default is

For a detailed description of the time-series plot, see the Time-Series Plot section
on page 168 and Schafer (1997, pp. 120126).

SAS OnlineDoc: Version 8

MONOTONE Statement

149

WLF

displays the worst linear function of parameters. This scalar function of parameters
 and  is worst in the sense that its values from iterations converge most slowly
among parameters. For a detailed description of this statistic, see the Worst Linear
Function of Parameters section on page 168.

MONOTONE Statement
MONOTONE < options > ;
The MONOTONE statement specifies an imputation method for data sets with monotone missingness. You must also specify a VAR statement and the data set must have a
monotone missing pattern with variables ordered in the VAR list. When both MONOTONE and MCMC statements are specified, the MONOTONE statement is not used..
You can specify the following options in a MONOTONE statement.
METHOD=REG | REGRESSION
METHOD=PROPENSITY < = NGROUPS = number>

specifies the imputation method for a data set with a monotone missing pattern. You can specify either METHOD=REG, a parametric regression method, or
METHOD=PROPENSITY, a nonparametric method based on propensity scores.
The default is METHOD=REG.
When METHOD=PROPENSITY is specified, the MAXIMUM=, MINIMUM=, and
ROUND= options, which make the imputed values more consistent with the observed
variable values, are not applicable.
NGROUPS=number

specifies the number of groups based on propensity scores for METHOD=PROPENSITY.

The default is NGROUPS=5.
See the Regression Method for Monotone Missing Data section on page 157 for a
detailed description of the regression method, and the Propensity Score Method for
Monotone Missing Data section on page 158 for the propensity score method.

SAS OnlineDoc: Version 8

150 

Chapter 9. The MI Procedure

TRANSFORM Statement
TRANSFORM transform ( variables < / options >)
< : : : transform ( variables < / options >) > ;
The TRANSFORM statement lists the transformations and their associated variables
to be transformed. The options are transformation options that provide additional
information for the transformation.
The MI procedure assumes that the data are from a multivariate normal distribution
when either the regression method or the MCMC method is used. When some variables in a data set are clearly non-normal, it is useful to transform these variables to
conform to the multivariate normality assumption. With a TRANSFORM statement,
variables are transformed before the imputation process and these transformed variable values are displayed in all of the results. When you specify an OUT= option, the
variable values are reverse-transformed to create the imputed data set.
The following transformations can be used as the transform in the TRANSFORM
statement.
BOXCOX

specifies the Box-Cox transformation of variables. The variable Y is transformed to

(Y+c) 1
, where c is a constant such that each value of Y + c must be positive and

the constant  > 0.
EXP

specifies the exponential transformation of variables. The variable Y is transformed

to e(Y+c) , where c is a constant.
LOG

specifies the logarithmic transformation of variables. The variable Y is transformed

to log(Y + c), where c is a constant such that each value of Y+c must be positive.
LOGIT

specifies the logit transformation of variables. The variable Y is transformed to

log( 1 YY=c=c ), where the constant c > 0 and the values of Y=c must be between 0
and 1.
POWER

specifies the power transformation of variables. The variable Y is transformed to

(Y + c) , where c is a constant such that each value of Y + c must be positive and
the constant  6= 0.

SAS OnlineDoc: Version 8

VAR Statement

151

The following options provide the constant c and  values in the transformations.
C=number

specifies the c value in the transformation. The default is c = 1 for logit transformation and c = 0 for other transformations.

LAMBDA=number

specifies the  value in the power and Box-Cox transformations. You must specify
the  value for these two transformations.
For example, the statement
transform log(y1) power(y2/c=1 lambda=.5);

requests that variables log(y1), a logarithmic transformation for the variable y1, and
py2 + 1, a power transformation for the variable y2, be used in the imputation.
If the MU0= option is used to specify a parameter value 0 for a transformed variable,
the same transformation for the variable is also applied to its corresponding MU0=
value in the t test. Otherwise, 0 = 0 is used for the transformed variable. See
Example 9.7 for a usage of the TRANSFORM statement.

VAR Statement
VAR variables ;
The VAR statement lists the variables to be analyzed. The variables must be numeric. If you omit the VAR statement, all numeric variables not mentioned in other
statements are used. The VAR statement is required if you specify a MONOTONE
statement, an IMPUTE=MONOTONE option in the MCMC statement, or more than
one number in the MU0=, MAXIMUM=, MINIMUM=, or ROUND= option.

SAS OnlineDoc: Version 8

152 

Chapter 9. The MI Procedure

Details
Descriptive Statistics
Suppose Y is the np matrix of complete data, which may not be fully observed,
n0 is the number of observations fully observed, and nj is the number of observations
with observed values for variable Yj .
With complete cases, the sample mean vector is

X
y = n1
yi
0

and the CSSCP matrix is

(yi y)(yi y)0

where each summation is over the fully observed observations.

The sample covariance matrix is

S= n 1 1
0

(yi y)(yi y)0

and is an unbiased estimate of the covariance matrix.

The correlation matrix R containing the Pearson product-moment correlations of the
variables is derived by scaling the corresponding covariance matrix:

R = D 1S D

where D is a diagonal matrix whose diagonal elements are the square roots of the
diagonal elements of S.
With available cases, the corrected sum of squares for variable Yj is

(yji yj )2
P

where y j = n1j
yji is the sample mean and each summation is over observations
with observed values for variable Yj .
The variance is

s2jj =

nj

(yji yj )2

The correlations for available cases contain pairwise correlations for each pair of
variables. Each correlation is computed from all observations that have nonmissing
values for the corresponding pair of variables.
SAS OnlineDoc: Version 8

EM Algorithm for Data with Missing Values

153

EM Algorithm for Data with Missing Values

The EM algorithm (Dempster, Laird, and Rubin 1977) is a technique that finds maximum likelihood estimates in parametric models for incomplete data. The books by
Little and Rubin (1987), Schafer (1997), and McLachlan and Krishnan (1997) provide detailed description and applications of the EM algorithm.
The EM algorithm is an iterative procedure that finds the MLE of the parameter vector
by repeating the following steps:
1. The expectation E-step:
Given a set of parameter estimates, such as a mean vector and covariance matrix for a
multivariate normal distribution, the E-step calculates the conditional expectation of
the complete-data log likelihood given the observed data and the parameter estimates.
2. The maximization M-step:
Given a complete-data log likelihood, the M-step finds the parameter estimates to
maximize the complete-data log likelihood from the E-step.
The two steps are iterated until the iterations converge.
In the EM process, the observed-data log likelihood is non-decreasing at each iteration. For multivariate normal data, suppose there are G groups with distinct missing
patterns. Then the observed-data log likelihood being maximized can be expressed
as

ln L(jYobs ) =

G
X
g=1

ln Lg ( jYobs )

where ln Lg ( jYobs ) is the observed-data log likelihood from the gth group, and

ln Lg (jYobs ) = n2g ln jg j

1 X (y  )0  1 (y  )
2 ig ig g g ig g

where ng is the number of observations in the gth group, the summation is over
observations in the gth group, yig is a vector of observed values corresponding to
observed variables, g is the corresponding mean vector, and g is the associated
covariance matrix.
Refer to Schafer (1997, pp. 163181) for a detailed description of the EM algorithm
for multivariate normal data.
PROC MI uses the means and standard deviations from available cases as the initial
estimates for the EM algorithm. The correlations are set to zero. For a discussion of
suggested starting values for the algorithm, see Schafer (1997, p. 169).
You can specify the convergence criterion with the CONVERGE= option in the EM
statement. The iterations are considered to have converged when the maximum
change in the parameter estimates between iteration steps is less than the value specified. You can also specify the maximum number of iterations used in the EM algorithm with the MAXITER= option.
SAS OnlineDoc: Version 8

154 

Chapter 9. The MI Procedure

The MI procedure displays tables of the initial parameter estimates used to begin
the EM process and the MLE parameter estimates derived from EM. You can also
display the EM iteration history with the option ITPRINT. PROC MI lists the iteration
number, the likelihood -2 Log L, and parameter values  at each iteration. You can
also save the MLE derived from the EM algorithm in a SAS data set specified with
the OUTEM= option.

Statistical Assumptions for Multiple Imputation

The MI procedure assumes that the data are from a continuous multivariate distribution and contain missing values that can occur on any of the variables. It also assumes
that the data are from a multivariate normal distribution when either the regression
method or the MCMC method is used.
Suppose Y is the np matrix of complete data, which is not fully observed, and
denote the observed part of Y by Yobs and the missing part by Ymis . The SAS MI
and MIANALYZE procedures assume that the missing data are missing at random
(MAR), that is, the probability that an observation is missing can depend on Yobs ,
but not on Ymis (Rubin 1976; 1987, p. 53).
To be more precise, suppose that R is the np matrix of response indicators whose
elements are zero or one depending on whether the corresponding elements of Y are
missing or observed. Then the MAR assumption is that the distribution of R can
depend on Yobs but not on Ymis .

p(RjYobs ; Ymis ) = p(RjYobs )

For example, consider a trivariate data set with variables Y1 and Y2 fully observed,
and a variable Y3 that has missing values. MAR assumes that the probability that
Y3 is missing for an individual can be related to the individuals values of variables
Y1 and Y2 , but not to its value of Y3 . On the other hand, if a complete case and an
incomplete case for Y3 with exactly the same values for variables Y1 and Y2 have
systematically different values, then there exists a response bias for Y3 , and MAR is
violated.
The MAR assumption is not the same as missing completely at random (MCAR),
which is a special case of MAR. Under the MCAR assumption, the missing data
values are a simple random sample of all data values; the missingness does not depend
on the values of any variables in the data set.
Furthermore, the MI and MIANALYZE procedures assume that the parameters  of
the data model and the parameters  of the model for the missing data indicators are
distinct. That is, knowing the values of  does not provide any additional information
about , and vice versa. If both the MAR and distinctness assumptions are satisfied,
the missing-data mechanism is said to be ignorable (Rubin 1987, pp. 5054; Schafer
1997, pp. 1011) .

SAS OnlineDoc: Version 8

Missing Data Patterns

155

Missing Data Patterns

The MI procedure sorts the data into groups based on whether an individuals value
is observed or missing for each variable to be analyzed. The input data set does not
need to be sorted in any order.
For example, with variables Y1 , Y2 , and Y3 (in that order) in a data set, up to eight
groups of observations can be formed from the data set. The following figure displays
the eight groups of observations and an unique missing pattern for each group:
Missing Data Patterns

Figure 9.6.

Group

Y1

Y2

Y3

1
2
3
4
5
6
7
8

X
X
X
X
.
.
.
.

X
X
.
.
X
X
.
.

X
.
X
.
X
.
X
.

Missing Data Patterns

Here, an X means that the variable is observed in the corresponding group and a
. means that the variable is missing.
The variable order is used to derive the order of the groups from the data set, and thus
determines the order of missing values in the data to be imputed. If you specify a
different order of variables in the VAR statement, then the results are different even
if the other specifications remain the same.
A data set with variables Y1 , Y2 , ..., Yp (in that order) is said to have a monotone
missing pattern when the event that a variable Yj is missing for a particular individual implies that all subsequent variables Yk , k > j , are missing for that individual.
Alternatively, when a variable Yj is observed for a particular individual, it is assumed
that all previous variables Yk , k < j , are also observed for that individual.
For example, the following figure displays a data set of three variables with a monotone missing pattern. Note that this data set does not have any observations with
missing patterns such as in Groups 3, 5, 6, 7, or 8 in the previous example.
Monotone Missing Data Patterns

Figure 9.7.

Group

Y1

Y2

Y3

1
2
3

X
X
X

X
X
.

X
.
.

Monotone Missing Patterns

SAS OnlineDoc: Version 8

156 

Chapter 9. The MI Procedure

Imputation Mechanisms
This section describes the three methods for multiple imputation that are available in
the MI procedure. The method of choice depends on the patterns of missingness in
the data.




For data sets with monotone missing patterns, either a parametric regression
method (Rubin 1987) that assumes multivariate normality or a nonparametric
method that uses propensity scores (Rubin 1987; Lavori, Dawson, and Shera
1995) is appropriate.
For data sets with arbitrary missing patterns, a Markov Chain Monte Carlo
(MCMC) method (Schafer 1997) that assumes multivariate normality is used
to impute either all missing values or just enough missing values to make the
imputed data sets have monotone missing patterns.

With a monotone missing data pattern, you have greater flexibility in your choice of
strategies. For example, in addition to the MCMC method, you can also implement
other methods, such as a regression method, that do not use Markov chains.
With an arbitrary missing data pattern, you can often use the MCMC method, which
creates multiple imputations by drawing simulations from a Bayesian predictive distribution for normal data. Another way to handle a data set with an arbitrary missing data pattern is to use the MCMC approach to impute enough values to make
the missing data pattern monotone. Then, you can use a more flexible imputation
method. This approach is described in the Producing Monotone Missingness with
the MCMC Method section on page 164.
Although the regression and MCMC methods assume multivariate normality, inferences based on multiple imputation can be robust to departures from the multivariate
normality if the amount of missing information is not large. It often makes sense
to use a normal model to create multiple imputations even when the observed data
are somewhat non-normal, as supported by simulation studies described in Schafer
(1997) and the original references therein.
You can also use a TRANSFORM statement to transform variables to conform to the
multivariate normality assumption. With a TRANSFORM statement, variables are
transformed before the imputation process and then are reverse-transformed to create
the imputed data set.
Li (1988) presented an argument for convergence of the MCMC method in the continuous case in theory and used it to create imputations for incomplete multivariate
continuous data. But in practice, it is not easy to check the convergence of a Markov
chain, especially for parameters from a large number of variables. PROC MI generates statistics and plots that you can use to check for convergence of the MCMC process. The details are described in the Convergence in MCMC section on page 167.

SAS OnlineDoc: Version 8

Regression Method for Monotone Missing Data

157

Regression Method for Monotone Missing Data

A data set with variables Y1 , Y2 , ..., Yp (in that order) is said to have a monotone missing pattern when the event that a variable Yj is observed for a particular individual
implies that all previous variables Yk , k < j , are also observed for that individual.
In the regression method, a regression model is fitted for each variable with missing
values, with the previous variables as covariates. Based on the fitted regression coefficients, a new regression model is simulated from the posterior predictive distribution
of the parameters and is used to impute the missing values for each variable (Rubin
1987, pp. 166167). The process is repeated sequentially for variables with missing
values. That is, for a variable Yj with missing values, a model

Yj = 0 + 1 Y1 + 2 Y2 + : : : + j 1 Yj

is fitted using observations with observed values for variables Y1 , Y2 , ..., Yj .

The fitted model includes the regression parameter estimates ^ = (^0 ; ^1 ; :::; ^j 1 )

and the associated covariance matrix 
^j2 Vj , where Vj is the usual X0 X inverse matrix derived from the intercept and variables Y1 ; Y2 ; :::; Yj 1 .
For each imputation, new parameters  = (0 ; 1 ; :::; (j 1) ) and 2j are drawn
from the posterior predictive distribution of the parameters. That is, they are simulated from (^0 ; ^1 ; :::; ^j 1 ), j2 , and Vj . The variance is drawn as

2j = ^j2 (nj

j )=g

where g is a 2nj j random variate and nj is the number of nonmissing observations

for Yj . The regression coefficients are drawn as

0 Z
 = ^ + j Vhj
0 is the upper triangular matrix in the Cholesky decomposition,
where Vhj
0 Vhj , and Z is a vector of j independent random normal variates.
Vj = Vhj
The missing values are then replaced by

0 + 1 y1 + 2 y2 + : : : + (j

1)

yj

+ zi j

where y1 ; y2 ; :::; yj 1 are the covariate values of the first j

simulated normal deviate.

1 variables and zi is a

SAS OnlineDoc: Version 8

158 

Chapter 9. The MI Procedure

Propensity Score Method for Monotone Missing Data

A propensity score is generally defined as the conditional probability of assignment
to a particular treatment given a vector of observed covariates (Rosenbaum and Rubin 1983). In the propensity score method, for each variable with missing values, a
propensity score is generated for each observation to estimate the probability that the
observation is missing. The observations are then grouped based on these propensity
scores, and an approximate Bayesian bootstrap imputation (Rubin 1987, p. 124) is
applied to each group (Lavori, Dawson, and Shera 1995).
A data set with variables Y1 , Y2 , ..., Yp (in that order) is said to have a monotone missing pattern when the event that a variable Yj is observed for a particular individual
implies that all previous variables Yk , k < j , are also observed for that individual.
The propensity score method uses the following steps to impute values for each variable Yj with missing values:
1. Create an indicator variable
and 1 otherwise.

Rj with the value 0 for observations with missing Yj

2. Fit a logistic regression model

logit(pj ) = 0 + 1 Y1 + 2 Y2 + : : : + j 1 Yj
where pj

= P r(Rj = 0jY1 ; Y2 ; :::; Yj 1 )

and

logit(p) = log(p=(1 p)):

3. Create a propensity score for each observation to estimate the probability that it is
missing.
4. Divide the observations into a fixed number of groups (typically assumed to be
five) based on these propensity scores.
5. Apply an approximate Bayesian bootstrap imputation to each group. In group k ,
suppose that Yobs denotes the n1 observations with nonmissing Yj values and Ymis
denotes the n0 observations with missing Yj . The approximate Bayesian bootstrap
imputation first draws n1 observations randomly with replacement from Yobs to create
 . This is a nonparametric analogue of drawing parameters from
a new data set Yobs
the posterior predictive distribution of the parameters. The process then draws the n0
 .
values for Ymis randomly with replacement from Yobs
Steps 1 through 5 are repeated sequentially for each variable with missing values.
Note that the propensity score method was originally designed for a randomized experiment with repeated measures on the response variables. The goal was to impute
the missing values on the response variables. The method uses only the covariate
information that is associated with whether the imputed variable values are missing.
It does not use correlations among variables. It is effective for inferences about the
distributions of individual imputed variables, such as an univariate analysis, but it is
not appropriate for analyses involving relationship among variables, such as a regression analysis. It can also produce badly biased estimates of regression coefficients
when data on predictor variables are missing (Allison 2000).

SAS OnlineDoc: Version 8

MCMC Method for Arbitrary Missing Data

159

MCMC Method for Arbitrary Missing Data

The Markov Chain Monte Carlo (MCMC) method originated in physics as a tool
for exploring equilibrium distributions of interacting molecules. In statistical applications, it is used to generate pseudo-random draws from multidimensional and
otherwise intractable probability distributions via Markov chains. A Markov chain
is a sequence of random variables in which the distribution of each element depends
only on the value of the previous one.
In MCMC simulation, one constructs a Markov chain long enough for the distribution
of the elements to stabilize to a stationary distribution, which is the distribution of
interest. By repeatedly simulating steps of the chain, the method simulates draws
from the distribution of interest. Refer to Schafer (1997) for a detailed discussion of
this method.
In Bayesian inference, information about unknown parameters is expressed in the
form of a posterior probability distribution. This posterior distribution is computed
using Bayes theorem

p(yj )p()
p(jy) = R
p(yj)p()d
MCMC has been applied as a method for exploring posterior distributions in Bayesian
inference. That is, through MCMC, one can simulate the entire joint posterior distribution of the unknown quantities and obtain simulation-based estimates of posterior
parameters that are of interest.
In many incomplete data problems, the observed-data posterior p( jYobs ) is intractable and cannot easily be simulated. However, when Yobs is augmented by
an estimated/simulated value of the missing data Ymis , the complete-data posterior
p(jYobs ; Ymis ) is much easier to simulate. Assuming that the data are from a multivariate normal distribution, data augmentation can be applied to Bayesian inference
with missing data by repeating the following steps:
1. The imputation I-step:
Given an estimated mean vector and covariance matrix, the I-step simulates the missing values for each observation independently. That is, if you denote the variables
with missing values for observation i by Yi(mis) and the variables with observed values by Yi(obs) , then the I-step draws values for Yi(mis) from a conditional distribution
for Yi(mis) given Yi(obs) .
2. The posterior P-step:
Given a complete sample, the P-step simulates the posterior population mean vector
and covariance matrix. These new estimates are then used in the next I-step. Without
prior information about the parameters, a noninformative prior is used. You can also
use other informative priors. For example, a prior information about the covariance
matrix can be helpful to stabilize the inference about the mean vector for a near
singular covariance matrix.

SAS OnlineDoc: Version 8

160 

Chapter 9. The MI Procedure

The two steps are iterated long enough for the results to be reliable for a multiply
imputed data set (Schafer 1997, p. 72). That is, with a current parameter estimate
(t+1)
(t) at the tth iteration, the I-step draws Ymis
from p(Ymis jYobs ;  (t) ) and the P-step
(t+1)
draws  (t+1) from p( jYobs ; Ymis ).
This creates a Markov chain
(1)
(2)
(Ymis
;  (1) ) , (Ymis
; (2) ) , ... ,

which converges in distribution to p(Ymis ;  jYobs ). Assuming the iterates converge to

a stationary distribution, the goal is to simulate an approximately independent draw
of the missing values from this distribution.
To validate the imputation results, you should repeat the process with different random number generators and starting values based on different initial parameter estimates.
The next three sections provide details for the imputation step, Bayesian estimation
of the mean vector and covariance matrix, and the posterior step.

Imputation Step
In each iteration, starting with a given mean vector  and covariance matrix , the
imputation step draws values for the missing data from the conditional distribution
Ymis given Yobs .
Suppose  = [01 ; 02 ]0 is the partitioned mean vector of two sets of variables, Yobs
and Ymis , where 1 is the mean vector for variables Yobs and 2 is the mean vector
for variables Ymis .
Also suppose

 =

11 12
012 22

is the partitioned covariance matrix for these variables, where 11 is the covariance
matrix for variables Yobs , 22 is the covariance matrix for variables Ymis , and 12 is
the covariance matrix between variables Yobs and variables Ymis .
By using the sweep operator (Goodnight 1979) on the pivots of the 11 submatrix,
the matrix becomes

111 11112
0
12 111
22:1

where 22:1 = 22 012 111 12 can be used to compute the conditional covariance
matrix of Ymis after controlling for Yobs .

SAS OnlineDoc: Version 8

MCMC Method for Arbitrary Missing Data

161

For an observation with the preceding missing pattern, the conditional distribution of
Ymis given Yobs = y1 is a multivariate normal distribution with the mean vector

2:1 = 2 + 012 111 (y1 1 )

and the conditional covariance matrix

22:1 = 22 012 111 12

Bayesian Estimation of the Mean Vector and Covariance Matrix
Suppose that Y = ( y10 ; y20 ; :::; yn0 )0 is an (np) matrix made up of n (p1) independent vectors yi , each of which has a multivariate normal distribution with mean
zero and covariance matrix . Then the SSCP matrix

A = Y0Y =

X
i

yi yi0

has a Wishart distribution W (n; ).

When each observation yi is distributed with a multivariate normal distribution with
an unknown mean , then the CSSCP matrix

A=

X
i

(yi y)(yi y)0

has a Wishart distribution W (n

1; ).
If A has a Wishart distribution W (n; ), then B = A 1 has an inverted Wishart
distribution W 1 (n; ), where n is the degrees of freedom and =  1 is the
precision matrix (Anderson 1984).
Note that, instead of using the parameter = 
tion, Schafer (1997) uses the parameter .

for the inverted Wishart distribu-

Suppose that each observation in the data matrix Y has a multivariate normal distribution with mean  and covariance matrix . Then with a prior inverted Wishart
distribution for  and a prior normal distribution for 

 
j 

( m; )
1
N 0 ; 

W

where  > 0 is a fixed number. The posterior distribution (Anderson 1984, p. 270;
Schafer 1997, p. 152) is

jY 
j(; Y) 
where (n


1

n + m; (n 1)S + +

n

n+


(y 0 )(y

0 )0

1 (ny +   ); 1 
N
0
n+
n+

1)S is the CSSCP matrix.

SAS OnlineDoc: Version 8

162 

Chapter 9. The MI Procedure

Posterior Step
In each iteration, the posterior step simulates the posterior population mean vector
 and covariance matrix  from prior information for  and , and the complete
sample estimates.
You can specify the prior parameter information using one of the following methods:





PRIOR=JEFFREYS, which uses a noninformative prior.

PRIOR=INPUT=, which provides a prior information for  in the data set.
Optionally, it also provides a prior information for  in the data set.
PRIOR=RIDGE=, which uses a ridge prior.

The next four subsections provide details of the posterior step for different prior distributions.

1. A Noninformative Prior
Without prior information about the mean and covariance estimates, a noninformative prior can be used by specifying the PRIOR=JEFFREYS option. The posterior
distributions (Schafer 1997, p. 154) are

(t+1) jY 

( n 1; (n 1)S)
1
N y ; (t+1)
n
W

(t+1) j((t+1) ; Y) 

2. An Informative Prior for  and 

When prior information is available for the parameters  and , you can provide it
with a SAS data set that you specify with the PRIOR=INPUT= option.

 
j 

( d ; d S )
1
N 0 ; 
n
W

To obtain the prior distribution for , PROC MI reads the matrix S from observations in the data set with TYPE =COV, and it reads n = d + 1 from observations with TYPE =N.
To obtain the prior distribution for , PROC MI reads the mean vector
from observations with TYPE =MEAN, and it reads n0 from observations with TYPE =N MEAN. When there are no observations with
TYPE =N MEAN, PROC MI reads n0 from observations with TYPE =N.

0

SAS OnlineDoc: Version 8

MCMC Method for Arbitrary Missing Data

163

The resulting posterior distribution, as described in the Bayesian Estimation of the

Mean Vector and Covariance Matrix section on page 161, is given by

(t+1) jY 

(t+1) j (t+1) ; Y

W
N

( n + d ; (n 1)S + d S + Sm ) 
1 (ny + n  ); 1 (t+1)
0 0
n + n0
n + n0

where
0
0
Sm = nnn
+ n (y 0 )(y 0 )
0

3. An Informative Prior for 

When the sample covariance matrix S is singular or near singular, prior information
about  can also be used without prior information about  to stabilize the inference about . You can provide it with a SAS data set that you specify with the
PRIOR=INPUT= option.
To obtain the prior distribution for , PROC MI reads the matrix S from observations in the data set with TYPE =COV, and it reads n from observations with
TYPE =N.
Note that if the PRIOR=INPUT= data set also contains observations with
TYPE =MEAN, then a complete informative prior for both  and  will be
used.
Corresponding to the prior for 

 

( d ; d S )

the posterior distribution for  (Anderson 1984, p. 269) is

(t+1) jY 

( (n 1) + d ; (n 1)S + d S )

Thus, an estimate of  is given by the weighted average

1
 
(n 1) + d ((n 1)S + d S )

and the posterior distribution for (; ) becomes

(t+1) jY 

(t+1) j (t+1) ; Y

( (n 1) + d; (n 1)S + d S )
1 (t+1)
N y ;
n
W

SAS OnlineDoc: Version 8

164 

Chapter 9. The MI Procedure

4. A Ridge Prior
A special case of the preceding adjustment is a ridge prior with S = Diag S (Schafer
1997, p. 156). That is, S is a diagonal matrix with diagonal elements equal to the
corresponding elements in S.
You can request a ridge prior by using the PRIOR=RIDGE= option. You can explicitly specify the number d  1 in the PRIOR=RIDGE=d option. Or you can implicitly specify the number by specifying the proportion p in the PRIOR=RIDGE=p
option to request d = (n 1)p.
The posterior is then given by

(t+1) jY 

(t+1) j (t+1) ; Y

( (n 1) + d ; (n 1)S + d S )
1 (t+1)
N y;
n

Producing Monotone Missingness with the MCMC Method

The monotone data MCMC method was first proposed by Li (1988), and Liu (1993)
described the algorithm. The method is useful especially when a data set is close to
having a monotone missing pattern. In this case, the method only needs to impute a
few missing values to the data set to have a monotone missing pattern in the imputed
data set. Compared to a full data imputation that imputes all missing values, the
monotone data MCMC method imputes fewer missing values in each iteration and
achieves approximate stationarity in fewer iterations (Schafer 1997, p. 227).
You can request the monotone MCMC method by specifying the option
IMPUTE=MONOTONE in the MCMC statement. The Missing Data Patterns table
now denotes the variables with missing values by . or O. A . means that the
variable is missing and will be imputed and an O means that the variable is missing
and will not be imputed. The tables of Multiple Imputation Variance Information
and Multiple Imputation Parameter Estimates are not created.
You must specify the variables in the VAR statement. The variable order in the list
determines the monotone missing pattern in the imputed data set. With a different
order in the VAR list, the results will be different because the monotone missing
pattern to be constructed will be different.
Assuming that the data are from a multivariate normal distribution, then similar to the
MCMC method, the monotone MCMC method repeats the following steps:
1. The imputation I-step:
Given an estimated mean vector and covariance matrix, the I-step simulates the missing values for each observation independently. Only a subset of missing values are
simulated to achieve a monotone pattern of missingness.
2. The posterior P-step:
Given a new sample with a monotone pattern of missingness, the P-step simulates
the posterior population mean vector and covariance matrix with a noninformative
Jeffreys prior. These new estimates are then used in the next I-step.

SAS OnlineDoc: Version 8

Producing Monotone Missingness with the MCMC Method

165

Imputation Step
The I-step is almost identical to the I-step described in the MCMC Method for Arbitrary Missing Data section on page 159 except that here only a subset of missing
values need to be simulated. To state this precisely, denote the variables with observed values for observation i by Yi(obs) and the variables with missing values by
Yi(mis) = (Yi(m1) Yi(m2) ), where Yi(m1) is a subset of the the missing variables that
will result a monotone missingness when their values are imputed. Then the I-step
draws values for Yi(m1) from a conditional distribution for Yi(m1) given Yi(obs) .
Posterior Step
The P-step is different from the P-step described in the MCMC Method for Arbitrary
Missing Data section on page 159. Instead of simulating the  and  parameters
from the full imputed data set, the P-step here simulates the  and  parameters
through simulated regression coefficients from regression models based on the imputed data set with a monotone pattern of missingness. The step is similar to the
process described in the Regression Method for Monotone Missing Data section
on page 157.
That is, for the variable Yj , a model

Yj = 0 + 1 Y1 + 2 Y2 + : : : + j 1 Yj

is fitted using nonmissing observations.

^ =
The fitted model consists of the regression parameter estimates 
2
(^0 ; ^1 ; : : : ; ^j 1 ) and the associated covariance matrix ^j Vj , where Vj is the
usual X0 X inverse matrix from the intercept and variables Y1 ; Y2 ; :::; Yj 1 .
For each imputation, new parameters   = (0 ; 1 ; :::; (j 1) ) and 2j are drawn
from the posterior predictive distribution of the parameters. That is, they are simulated from (^0 ; ^1 ; :::; ^j 1 ), j2 , and Vj . The variance is drawn as

2j = ^j2 (nj

j )=g

where g is a 2nj p+j 1 random variate and nj is the number of nonmissing observations for Yj . The regression coefficients are drawn as

0 Z
 = ^ + j Vhj
0 is the upper triangular matrix in the Cholesky decomposition
where Vhj
0
Vj = Vhj Vhj and Z is a vector of j independent random normal variates.
These simulated values of   and 2j are then used to re-create the parameters 
and . For a detailed description of how to produce monotone-missingness with the
MCMC method for a multivariate normal data, refer to Schafer (1997, pp. 226235).

SAS OnlineDoc: Version 8

166 

Chapter 9. The MI Procedure

MCMC Method Specifications

With MCMC, you can impute either all missing values (IMPUTE=FULL) or just
enough missing values to make the imputed data set have a monotone missing pattern
(IMPUTE=MONOTONE). In the process, either a single chain for all imputations
(CHAIN=SINGLE) or a separate chain for each imputation (CHAIN=MULTIPLE)
is used. Refer to Schafer (1997, pp. 137138) for a discussion of single versus
multiple chains.
You can specify the number of initial burn-in iterations before the first imputation
with the NBITER= option. This number is also used for subsequent chains for multiple chains. For a single chain, you can also specify the number of iterations between
imputations with the NITER= option.
You can explicitly specify initial parameter values for the MCMC process with the
INITIAL=INPUT= data set option. Alternatively, you can use the EM algorithm to
derive a set of initial parameter values for MCMC with the option INITIAL=EM.
These estimates are used as either the starting value (START=VALUE) or as the
starting distribution (START=DIST) for the MCMC process. For multiple chains,
these estimates are used again as either the starting value (START=VALUE) or as the
starting distribution (START=DIST) for the subsequent chains.
You can specify the prior parameter information in the PRIOR= option. You can use
a noninformative prior (PRIOR=JEFFREYS), a ridge prior (PRIOR=RIDGE), or an
informative prior specified in a data set (PRIOR=INPUT).
The parameter estimates used to generate imputed values in each imputation can be
saved in a data set with the OUTEST= option. Later, this data set can be read with
the INEST= option to provide the reference distribution for imputing missing values
for a new data set.
By default, the MCMC method uses a single chain to produce five imputations. It
completes 200 burn-in iterations before the first imputation and 100 iterations between imputations. The posterior mode computed from the EM algorithm with a
noninformative prior is used as the starting values for the MCMC process.

INITIAL=EM Specifications
The EM algorithm is used to find the maximum likelihood estimates for incomplete
data in the EM statement. You can also use the EM algorithm to find a posterior
mode, the parameter estimates that maximize the observed-data posterior density.
The resulting posterior mode provides a good starting value for the MCMC process.
With INITIAL=EM, PROC MI uses the MLE of the parameter vector as the initial
estimates in the EM algorithm for the posterior mode. You can use the ITPRINT
option in INITIAL=EM to display the iteration history for the EM algorithm.
You can use the CONVERGE= option to specify the convergence criterion in deriving
the EM posterior mode. The iterations are considered to have converged when the
maximum change in the parameter estimates between iteration steps is less than the
value specified. By default, CONVERGE=1E-4.

SAS OnlineDoc: Version 8

Convergence in MCMC

167

You can also use the MAXITER= option to specify the maximum number of iterations in the EM algorithm. By default, MAXITER=200.
With the BOOTSTRAP option, you can use overdispersed starting values for the
MCMC process. In this case, PROC MI applies the EM algorithm to a bootstrap
sample, a simple random sample with replacement from the input data set, to derive
the initial estimates for each chain (Schafer 1997, p. 128).

Convergence in MCMC
The theoretical convergence of the MCMC process has been explored under various
conditions, as described in Schafer (1997, p. 70). However, in practice, verification
of convergence is not a simple matter and cannot be easily implemented in the MI
procedure.
The parameters used in the imputation step for each iteration can be saved in an output
data set with the OUTITER= option. These include the means, standard deviations,
covariances, the worst linear function, and observed-data LR statistics. You can then
monitor the convergence in a single chain by displaying time-series plots and autocorrelations for those parameter values (Schafer 1997, p. 120). The time-series and
autocorrelation function plots for parameters such as variable means, covariances,
and the worst linear function can be displayed by specifying the TIMEPLOT and
ACFPLOT option.
You can apply EM to a bootstrap sample to obtain overdispersed starting values for
multiple chains (Gelman and Rubin 1992). This provides a conservative estimate of
the number of iterations needed before each imputation.
The next four subsections provide useful statistics and plots that can be used to check
the convergence of the MCMC process.

LR Statistics
You can save the observed-data likelihood ratio (LR) statistic in each iteration with
the LR option in the OUTITER= data set. The statistic is based on the observeddata likelihood with parameter values used in the iteration and the observed-data
maximum likelihood derived from the EM algorithm.
In each iteration, the LR statistic is given by

^
2 log f (^i )
f ( )

^ ) is the observed-data maximum likelihood derived from the EM algorithm

where f (
^ i ) is the observed-data likelihood for ^ i used in the iteration.
and f (
Similarly, you can also save the observed-data LR posterior mode statistic for each
iteration with the LR POST option. This statistic is based on the observed-data posterior density with parameter values used in each iteration and the observed-data posterior mode derived from the EM algorithm for posterior mode.

SAS OnlineDoc: Version 8

168 

Chapter 9. The MI Procedure

For large samples, these LR statistics tends to be approximately 2 distributed with
degrees of freedom equal to the dimension of  (Schafer 1997, p. 131). For example,
with a large number of iterations, if the values of the LR statistic do not behave like
a random sample from the described 2 distribution, then there is evidence that the
MCMC process has not converged.

Worst Linear Function of Parameters

The worst linear function (WLF) of parameters (Schafer 1997, pp. 129-131) is a
scalar function of parameters  and  that is worst in the sense that its function
values converge most slowly among parameters in the MCMC process. The convergence of this function is evidence that other parameters are likely to converge as
well.
For linear functions of parameters  = (; ), a worst linear function of  has the
highest asymptotic rate of missing information. The function can be derived from
the iterative values of  near the posterior mode in the EM algorithm. That is, an
estimated worst linear function of  is

w( ) = v0 ( ^ )

^ is the posterior mode and the coefficients v = ^ ( 1) ^ is the difference

where 
between the estimated value of  one step prior to convergence and the converged
^.
value 
You can display the coefficients of the worst linear function, v, by specifying the
WLF option in the MCMC statement. You can save the function value from each
iteration in an OUTITER= data set by specifying the WLF option in the OUTITER
option. You can also display the worst linear function values from iterations in an
autocorrelation plot or a time-series plot by specifying WLF as an ACFPLOT or
TIMEPLOT option, respectively.
Note that when the observed-data posterior is nearly normal, the WLF is one of the
slowest functions to approach stationarity. When the posterior is not close to normal,
other functions may take much longer than the WLF to converge, as described in
Schafer (1997, p.130).

Time-Series Plot
A time-series plot for a parameter  is a scatter plot of successive parameter estimates
i against the iteration number i. The plot provides a simple way to examine the
convergence behavior of the estimation algorithm for  . Long-term trends in the plot
indicate that successive iterations are highly correlated and that the series of iterations
has not converged.
You can display time-series plots for the worst linear function, the variable means,
variable variances, and covariances of variables. You can also request logarithmic
transformations for positive parameters in the plots with the LOG option. When a
parameter value is less than or equal to zero, the value is not displayed in the corresponding plot.

SAS OnlineDoc: Version 8

Convergence in MCMC

169

By default, the MI procedure uses the plus sign (+) as the plot symbol to display the
points with a height of one (percentage screen unit) in a time-series plot. You can use
the SYMBOL=, CSYMBOL=, and HSYMBOL= options to change the shape, color,
and height of the plot symbol.
By default, the plot title Time-Series Plot is displayed in a time-series plot. You
can request another title by using the TITLE= option in TIMEPLOT. When another
title is also specified in a TITLE statement, this title is displayed as the main title and
the plot title is displayed as a subtitle in the plot.
You can use options in the GOPTIONS statement to change the color and height of the
title. Refer to the chapter The SAS/GRAPH Statements in SAS/GRAPH Software:
Reference, Version 8 for a description of title options. See Example 9.6 for a usage
of the time-series plot.

Autocorrelation Function Plot

To examine relationships of successive parameter estimates  , the autocorrelation
function (ACF) can be used. For a stationary series, i ; i  1, in time series data, the
autocorrelation function at lag k is

k =

Cov(i ; i+k )
Var(i )

The sample kth order autocorrelation is computed as

rk =

Pn

)(i+k  )
(  )2

k
i=1Pi 
n
i=1 i

You can display autocorrelation function plots for the worst linear function, the variable means, variable variances, and covariances of variables. You can also request
logarithmic transformations for parameters in the plots with the LOG option. When a
parameter has values less than or equal to zero, the corresponding plot is not created.
You specify the maximum number of lags of the series with the NLAG= option. The
autocorrelations at each lag less than or equal to the specified lag are displayed in the
graph. In addition, the plot also displays approximate 95% confidence limits for the
autocorrelations. At lag k , the confidence limits indicate a set of approximate 95%
critical values for testing the hypothesis j = 0; j  k:
By default, the MI procedure uses the star sign (*) as the plot symbol to display the
points with a height of one (percentage screen unit) in the plot, a solid line to display
the reference line of zero autocorrelation, vertical line segments to connect autocorrelations to the reference line, and a pair of dashed lines to display approximately 95%
confidence limits for the autocorrelations.
You can use the SYMBOL=, CSYMBOL=, and HSYMBOL= options to change the
shape, color, and height of the plot symbol, and the CNEEDLES= and WNEEDLES=
options to change the color and width of the needles. You can also use the LREF=,
CREF=, and WREF= options to change the line type, color, and width of the reference line. Similarly, you can use the LCONF=, CCONF=, and WCONF= options to
change the line type, color, and width of the confidence limits.
SAS OnlineDoc: Version 8

170 

Chapter 9. The MI Procedure

By default, the plot title Autocorrelation Plot is displayed in a autocorrelation function plot. You can request another title by using the TITLE= option in ACFPLOT.
When another title is also specified in a TITLE statement, this title is displayed as the
main title and the plot title is displayed as a subtitle in the plot.
You can use options in the GOPTIONS statement to change the color and height of the
title. Refer to the chapter The SAS/GRAPH Statements in SAS/GRAPH Software:
Reference, Version 8 for a description of title options. See Example 9.6 for a usage
of the autocorrelation function plot.

Input Data Sets

You can specify the input data set with missing values with the DATA= option in
the PROC MI statement. When a MCMC method is used, you can specify the data
set containing the reference distribution information for imputation with the INEST=
option, the data set containing initial parameter estimates for the MCMC process with
the INITIAL=INPUT= option, and the data set containing information for the prior
distribution with the PRIOR=INPUT= option in the MCMC statement.
DATA=SAS-data-set

The input DATA= data set is an ordinary SAS data set containing multivariate data
with missing values.
INEST=SAS-data-set

The input INEST= data set is a TYPE=EST data set and contains a variable
Imputation to identify the imputation number. For each imputation, PROC
MI reads the point estimate from the observations with TYPE =PARM or
TYPE =PARMS and the associated covariances from the observations with
TYPE =COV or TYPE =COVB. These estimates are used as the reference
distribution to impute values for observations in the DATA= data set. When the input INEST= data set also contains observations with TYPE =SEED, PROC MI
reads the seed information for the random number generator from these observations.
Otherwise, the SEED= option provides the seed information. See Example 9.8 for a
usage of this option.
INITIAL=INPUT=SAS-data-set

The input INITIAL=INPUT= data set is a TYPE=COV or CORR data set and provides initial parameter estimates for the MCMC process. The covariances derived
from the TYPE=COV/CORR data set are divided by the number of observations to
get the correct covariance matrix for the point estimate (sample mean).
If TYPE=COV, PROC MI reads the number of observations from the observations with TYPE =N, the point estimate from the observations
with TYPE =MEAN, and the covariances from the observations with
TYPE =COV.
If TYPE=CORR, PROC MI reads the number of observations from the observations with TYPE =N, the point estimate from the observations with
TYPE =MEAN, the correlations from the observations with TYPE =CORR,
and the standard deviations from the observations with TYPE =STD.

SAS OnlineDoc: Version 8

Output Data Sets

171

PRIOR=INPUT=SAS-data-set

The input PRIOR=INPUT= data set is a TYPE=COV data set that provides information for the prior distribution. You can use the data set to specify a prior distribution
for  of the form

  W 1 ( d ; d S )
where d = n 1 is the degrees of freedom. PROC MI reads the matrix S from
observations with TYPE =COV and n from observations with TYPE =N.
You can also use this data set to specify a prior distribution for  of the form

1
  N 0 ; 
n

PROC MI reads the mean vector 0 from observations with TYPE =MEAN
and n0 from observations with TYPE =N MEAN. When there are no observations with TYPE =N MEAN, PROC MI reads n0 from observations with
TYPE =N.

Output Data Sets

You can specify the output data set of imputed values with the OUT= option in the
PROC MI statement. When an EM statement is used, you can specify the data set
containing MLE computed with the EM algorithm with the OUTEM= option in the
EM statement. When a MCMC method is used, you can specify the data set containing parameter estimates used in each imputation with the OUTEST= option and the
data set containing parameters used in the imputation step for each iteration with the
OUTITER option in the MCMC statement.
OUT=SAS-data-set

The OUT= data set contains all the variables in the original data set and a new variable
named Imputation that identifies the imputation. For each imputation, the data set
contains all variables in the input DATA= data set with missing values replaced by
imputed values.
OUTEM=SAS-data-set

The OUTEM= data set is a TYPE=COV data set and contains the MLE computed
with the EM algorithm. The observations with TYPE =MEAN contain the estimated mean and the observations with TYPE =COV contain the estimated covariances.
OUTEST=SAS-data-set

The OUTEST= data set is a TYPE=EST data set and contains parameter estimates
used in each imputation in the MCMC method. It also includes an index variable
named Imputation , which identifies the imputation.

SAS OnlineDoc: Version 8

172 

Chapter 9. The MI Procedure

The observations with TYPE =SEED contain the seed information for
the random number generator.
The observations with TYPE =PARM
or TYPE =PARMS contain the point estimate and the observations with
TYPE =COV or TYPE =COVB contain the associated covariances. These
estimates are used as the parameters of the reference distribution to impute values for
observations in the DATA= dataset.
Note that these estimates are the values used in the I-step before each imputation.
These are not the parameter values simulated from the P-step in the same iteration.
See Example 9.8 for a usage of this option.

OUTITER < ( options ) > =SAS-data-set in an EM statement

The OUTITER= data set in an EM statement is a TYPE=COV data set and contains
parameters for each iteration. It also includes a variable Iteration that provides
the iteration number.
The parameters in the output data set depend on the options specified. You can
specify the MEAN and COV options for OUTITER. With the MEAN option, the
output data set contains the mean parameters in observations with the variable
TYPE =MEAN. Similarly, with the MEAN option, the output data set contains
the covariance parameters in observations with the variable TYPE =COV. When
no options are specified, the output data set contains the mean parameters for each
iteration.
OUTITER < ( options ) > =SAS-data-set in a MCMC statement

The OUTITER= data set in a MCMC statement is a TYPE=COV data set and contains
parameters used in the imputation step for each iteration. It also includes variables
named Imputation and Iteration , which provide the imputation number and
iteration number.
The parameters in the output data set depend on the options specified. The following
table summarizes the options available for OUTITER and the corresponding values
for the output variable TYPE .
Table 9.3.

Options
MEAN
STD
COV
LR
LR POST
WLF

Summary of Options for OUTITER in a MCMC statement

Output Parameters
mean parameters
standard deviations
covariances
-2 log LR statistic
-2 log LR statistic of the posterior mode
worst linear function

TYPE
MEAN
STD
COV
LOG LR
LOG POST
WLF

When no options are specified, the output data set contains the mean parameters used
in the imputation step for each iteration. For a detailed description of the worst linear
function and LR statistics, see the Convergence in MCMC section on page 167.

SAS OnlineDoc: Version 8

Combining Inferences from Multiply Imputed Data Sets

173

Combining Inferences from Multiply Imputed Data Sets

With m imputations, m different sets of the point and variance estimates for a parameter Q can be computed. Suppose Q^i and U^i are the point and variance estimates
from the ith imputed data set, i=1, 2, ..., m. Then the combined point estimate for Q
from multiple imputation is the average of the m complete-data estimates:

Q=

m
1X
Q^i
m
i=1

Suppose U is the within-imputation variance, which is the average of the m completedata estimates:

U=

m
1X
U^i
m
i=1

and B is the between-imputation variance

B=

m
1 X
(Q^ Q)2
m 1 i=1 i

Then the variance estimate associated with Q is the total variance (Rubin 1987)

T = U + (1 +

1 )B

The statistic (Q Q)T (1=2) is approximately distributed as

freedom (Rubin 1987), where

vm = (m 1)[1 +

t with vm degrees of

(1 + m 1 )B ]

When the complete-data degrees of freedom v0 is small, and there is only a modest
proportion of missing data, the computed degrees of freedom, vm , can be much larger
than v0 , which is inappropriate. Barnard and Rubin (1999) recommend the use of an
adjusted degrees of freedom

1
1
m= v +v
^obs
m

v

= (1 + m 1 )B=T .
 , for inference.
Note that the MI procedure uses the adjusted degrees of freedom, vm
The degrees of freedom vm depends on m and the ratio

where

v^obs = (1 ) v0 (v0 + 1)=(v0 + 3)

r=

and

(1 + m 1 )B
U

SAS OnlineDoc: Version 8

174 

Chapter 9. The MI Procedure

The ratio r is called the relative increase in variance due to nonresponse (Rubin 1987).
When there is no missing information about Q, the values of r and B are both zero.
With a large value of m or a small value of r , the degrees of freedom v will be large
and the distribution of (Q Q)T (1=2) will be approximately normal.
Another useful statistic is the fraction of missing information about Q:

^ =

r + 2=(v + 3)
r+1

Both statistics r and  are helpful diagnostics for assessing how the missing data
contribute to the uncertainty about Q.

Multiple Imputation Efficiency

The relative efficiency (RE) of using the finite m imputation estimator, rather than
using an infinite number for the fully efficient imputation, in units of variance, is
approximately a function of m and  (Rubin 1987, p. 114).

RE = (1 +


)
m

The following table shows relative efficiencies with different values of m and .
For cases with little missing information, only a small number of imputations are
necessary.
Table 9.4.

Relative Efficiency

3
5
10
20


10%

0:9677
0:9804
0:9901
0:9950

20%

0:9375
0:9615
0:9804
0:9901

30%

0:9091
0:9434
0:9709
0:9852

50%

0:8571
0:9091
0:9524
0:9756

70%

0:8108
0:8772
0:9346
0:9662

Imputers Model Versus Analysts Model

Schafer (1997, pp. 139-143) provides comprehensive coverage of this topic, and the
following discussion is largely based on his work.
Multiple imputation inference assumes that the model you used to analyze the multiply imputed data (the analysts model) is the same as the model used to impute
missing values in multiple imputation (the imputers model). But in practice, the two
models may not be the same.

SAS OnlineDoc: Version 8

Parameter Simulation Versus Multiple Imputation

175

For example, consider the same trivariate data set with variables Y1 and Y2 fully
observed, and a variable Y3 with missing values. An imputer creates multiple imputations with the model Y3 = Y1 Y2 . However, the analyst can later use the simpler
model Y3 = Y1 . In this case, the analyst assumes more than the imputer. That is, the
analyst assumes there is no relationship between variables Y3 and Y2 .
The effect of the discrepancy between the models depends on whether the analysts
additional assumption is true. If the assumption is true, the imputers model still
applies. The inferences derived from multiple imputations will still be valid, although
they may be somewhat conservative because they reflect the additional uncertainty of
estimating the relationship between Y3 and Y2 .
On the other hand, suppose that the analyst models Y3 = Y1 , and there is a relationship between variables Y3 and Y2 . Then the model Y3 = Y1 will be biased and is
inappropriate. Appropriate results can be generated only from appropriate analysts
models.
Another type of discrepancy occurs when the imputer assumes more than the analyst.
For example, suppose that an imputer creates multiple imputations with the model
Y3 = Y1 , but the analyst later fits a model Y3 = Y1 Y2 . When the assumption is true,
the imputers model is a correct model and the inferences still hold.
On the other hand, suppose there is a relationship between Y3 and Y2 . Imputations
created under the incorrect assumption that there is no relationship between Y3 and
Y2 will make the analysts estimate of the relationship biased toward zero. Multiple
imputations created under an incorrect model can lead to incorrect conclusions.
Thus, generally you should include as many variables as you can when doing multiple imputation. The precision you lose when you include unimportant predictors
is usually a relatively small price to pay for the general validity of analyses of the
resultant multiply imputed data set (Rubin 1996).
Note that it is good practice to include a description of the imputers model with
the multiply imputed data set. That way, the analysts will have information about
the variables involved in the imputation and which relationships among the variables
have been implicitly set to zero.

Parameter Simulation Versus Multiple Imputation

For many incomplete-data problems, simulation-based methods of parameter simulation and multiple imputation can be used to analyze the data. In parameter simulation,
you simulate random values of parameters from the observed-data posterior distribution and make simple inferences about these parameters (Schafer 1997, p. 89).
When a set of well-defined population parameters  are of interest, parameter simulation can be used to directly examine and summarize simulated values of  . This
usually requires a large number of iterations, and involves calculating appropriate
summaries of the resulting dependent sample of the iterates of the  . If only a small
set of parameters are involved, parameter simulation can be suitable (Schafer 1997).

SAS OnlineDoc: Version 8

176 

Chapter 9. The MI Procedure

In multiple imputation, the unknown missing data are replaced by multiple sets of
simulated values. Each complete data set is then analyzed by standard complete-data
methods. The variability among the results from these repeated analyses provides a
measure of the uncertainty due to missing data. Combining this between-imputation
variation with the ordinary within-imputation sample variation provides statistical
inference for the parameters of interest. Multiple imputation is suitable for analyses
that are more exploratory in nature.
Multiple imputation only requires a small number of imputations. Generating and
storing a few imputations can be more efficient than generating and storing a large
number of iterations for parameter simulation.
When fractions of missing information are low, methods that average over simulated
values of the missing data, as in multiple imputation, can be much more efficient than
methods that average over simulated values of  as in parameter simulation (Schafer
1997).

ODS Table Names

PROC MI assigns a name to each table it creates. You must use these names to
reference tables when using the Output Delivery System (ODS). These names are
listed in the following table. For more information on ODS, refer to the chapter
Using the Output Delivery System in the SAS/STAT Users Guide, Version 8.
Table 9.5.

ODS Tables Produced in PROC MI

ODS Table Name

ModelInfo
MissPattern
Transform

Description
Model information
Missing data patterns
Variable Transformations

Univariate
Corr
EMInitEst
EMEst
EMIter

Univariate statistics for available cases

Pairwise correlations for available cases
Initial parameter values for EM
MLE of the parameter vector from EM
EM iteration history for MLE

EMPIter

EM iteration history for posterior mode

EMPEst
EMWlf
MCMCInitEst

Posterior mode parameter values from EM

Coefficients of the worst linear function
Initial parameter estimates for MCMC

VarianceInfo

Between-imputation, within-imputation, and

total variances
Parameter estimates

ParmEst

SAS OnlineDoc: Version 8

Option

TRANSFORM
statement
SIMPLE
SIMPLE
EM statement
EM statement
ITPRINT in EM
statement
ITPRINT in
INITIAL=EM
INITIAL=EM
WLF
DISPLAYINIT
in MCMC

Example 9.1.

EM Algorithm for MLE

177

Examples
The following FitMono data set has a monotone missing data pattern and is used in
Example 9.2 with the propensity score method and in Example 9.3 with the regression
method. The FitMiss data set created in the Getting Started section is used in other
examples. Note that the original data set has been altered for these examples.
*----------------- Data on Physical Fitness -----------------*
| These measurements were made on men involved in a physical |
| fitness course at N.C. State University.
|
| Only selected variables of
|
| Oxygen (oxygen intake, ml per kg body weight per minute), |
| Runtime (time to run 1.5 miles in minutes), and
|
| RunPulse (heart rate while running) are used.
|
| Certain values were changed to missing for the analysis.
|
*------------------------------------------------------------*;
data FitMono;
input Oxygen RunTime RunPulse @@;
datalines;
44.609 11.37 178
45.313 10.07 185
54.297
8.65 156
59.571
.
.
49.874
9.22
.
44.811 11.63 176
45.681 11.95 176
49.091 10.85
.
39.442 13.08 174
60.055
8.63 170
50.541
.
.
37.388 14.03 186
44.754 11.12 176
47.273
.
.
51.855 10.33 166
49.156
8.95 180
40.836 10.95 168
46.672 10.00
.
46.774 10.25
.
50.388 10.08 168
39.407 12.63 174
46.080 11.17 156
45.441
9.63 164
54.625
8.92 146
45.118 11.08
.
39.203 12.88 168
45.790 10.47 186
50.545
9.93 148
48.673
9.40 186
47.920 11.50 170
47.467 10.50 170
;

Example 9.1. EM Algorithm for MLE

This example uses the EM algorithm to compute the maximum likelihood estimates
for the parameters of a multivariate normal distribution using data with missing values. The following statements invoke the MI procedure and request the EM algorithm
to compute the MLE for (; ) of a multivariate normal distribution from the input
data set FitMiss.
proc mi data=FitMiss seed=55417 simple nimpute=0;
em itprint outem=outem;
var Oxygen RunTime RunPulse;
run;

Note when you specify the option NIMPUTE=0, the missing values will not be imputed. The procedure generates the following output:
SAS OnlineDoc: Version 8

178 

Chapter 9. The MI Procedure

Output 9.1.1.

Model Information
The MI Procedure
Model Information

Data Set
Method
Multiple Imputation Chain
Initial Estimates for MCMC
Start
Prior
Number of Imputations
Number of Burn-in Iterations
Number of Iterations
Seed for random number generator

WORK.FITMISS
MCMC
Single Chain
EM Posterior Mode
Starting Value
Jeffreys
0
200
100
55417

The Model Information table describes the method and options used in the procedure.
Output 9.1.2.

Missing Data Patterns

The MI Procedure
Missing Data Patterns

Group
1
2
3
4
5

Oxygen

Run
Time

Run
Pulse

X
X
X
.
.

X
X
.
X
X

X
.
.
X
.

Freq

Percent

21
4
3
1
2

67.74
12.90
9.68
3.23
6.45

Missing Data Patterns

Group
1
2
3
4
5

-----------------Group Means---------------Oxygen
RunTime
RunPulse
46.353810
47.109500
52.461667
.
.

10.809524
10.137500
.
11.950000
9.885000

171.666667
.
.
176.000000
.

The Missing Data Patterns table lists distinct missing data patterns with corresponding frequencies and percents. Here, X means that the variable is observed
in the corresponding group and . means that the variable is missing. The table also
displays group-specific variable means.

SAS OnlineDoc: Version 8

Example 9.1.

EM Algorithm for MLE

179

With the SIMPLE option, the procedure displays simple descriptive univariate statistics for available cases in the Univariate Statistics table and correlations from pairwise available cases in the Pairwise Correlations table.
Output 9.1.3.

Univariate Statistics
The MI Procedure
Univariate Statistics

Variable

Mean

Std Dev

Minimum

Maximum

Oxygen
RunTime
RunPulse

28
28
22

47.11618
10.68821
171.86364

5.41305
1.37988
10.14324

37.38800
8.63000
148.00000

60.05500
14.03000
186.00000

Output 9.1.4.

Pairwise Correlations
The MI Procedure
Pairwise Correlations

Oxygen
RunTime
RunPulse

Oxygen

RunTime

RunPulse

1.000000000
-0.849118562
-0.343961742

-0.849118562
1.000000000
0.247258191

-0.343961742
0.247258191
1.000000000

With the EM statement, the procedure displays the initial parameter estimates for
EM.
Output 9.1.5.

Initial Parameter Estimates for EM

The MI Procedure
Initial Parameter Estimates for EM

_TYPE_

_NAME_

MEAN
COV
COV
COV

Oxygen
RunTime
RunPulse

Oxygen

RunTime

RunPulse

47.116179
29.301078
0
0

10.688214
0
1.904067
0

171.863636
0
0
102.885281

SAS OnlineDoc: Version 8

180 

Chapter 9. The MI Procedure

With the ITPRINT option, the EM (MLE) Iteration History table displays the iteration history for the EM algorithm.
Output 9.1.6.

EM (MLE) Iteration History

The MI Procedure
EM (MLE) Iteration History

_Iteration_

-2 Log L

Oxygen

RunTime

RunPulse

0
1
2
3
4
5
6
7
8
9
10

289.544782
263.549489
255.851312
254.616428
254.494971
254.483973
254.482920
254.482813
254.482801
254.482800
254.482800

47.116179
47.116179
47.139089
47.122353
47.111080
47.106523
47.104899
47.104348
47.104165
47.104105
47.104086

10.688214
10.688214
10.603506
10.571685
10.560585
10.556768
10.555485
10.555062
10.554923
10.554878
10.554864

171.863636
171.863636
171.538203
171.426790
171.398296
171.389208
171.385257
171.383345
171.382424
171.381992
171.381796

The procedure then displays the EM (MLE) parameter estimates, the maximum likelihood estimates for  and  of a multivariate normal distribution from the data set
FitMiss.
Output 9.1.7.

EM (MLE) Parameter Estimates

The MI Procedure
EM (MLE) Parameter Estimates

_TYPE_

_NAME_

MEAN
COV
COV
COV

Oxygen
RunTime
RunPulse

SAS OnlineDoc: Version 8

Oxygen

RunTime

RunPulse

47.104086
27.798014
-6.457929
-18.030790

10.554864
-6.457929
2.015491
3.516092

171.381796
-18.030790
3.516092
97.766559

Example 9.2.

Propensity Score Method

181

You can also output the EM (MLE) parameter estimates into an output data set with
the OUTEM= option. The following statements list the observations in the output
data set outem.
proc print data=outem;
title EM Estimates;
run;

Output 9.1.8.

EM Estimates
EM Estimates

Obs

_TYPE_

_NAME_

Oxygen

RunTime

RunPulse

1
2
3
4

MEAN
COV
COV
COV

Oxygen
RunTime
RunPulse

47.1041
27.7980
-6.4579
-18.0308

10.5549
-6.4579
2.0155
3.5161

171.382
-18.031
3.516
97.767

The output data set outem is a TYPE=COV data set. The observation with
TYPE =MEAN contains the MLE for the parameter  and the observations with
TYPE =COV contain the MLE for the parameter  of a multivariate normal
distribution from the data set FitMiss.

Example 9.2. Propensity Score Method

This example uses the propensity score method to impute missing values in a data set
with a monotone missing pattern. The following statements invoke the MI procedure
and request the propensity score method. The resulting data set is named outpscore.
proc mi data=FitMono seed=55417 simple out=outpscore;
monotone method=propensity;
var Oxygen RunTime RunPulse;
run;

Note that the VAR statement is required and the data set must have a monotone missing pattern with variables as ordered in the VAR statement. The procedure generates
the following output:
Output 9.2.1.

Model Information
The MI Procedure
Model Information

Data Set
Method
Number of Imputations
Number of Groups on Propensity
Seed for random number generator

WORK.FITMONO
Propensity
5
5
55417

SAS OnlineDoc: Version 8

182 

Chapter 9. The MI Procedure

The Model Information table describes the method and options used in the multiple
imputation process. By default, the observations are sorted into five groups based on
the propensity scores, and five imputations are created for the missing data.
Output 9.2.2.

Missing Data Patterns

The MI Procedure
Missing Data Patterns

Group
1
2
3

Oxygen

Run
Time

Run
Pulse

X
X
X

X
X
.

X
.
.

Freq

Percent

23
5
3

74.19
16.13
9.68

Missing Data Patterns

-----------------Group Means---------------Oxygen
RunTime
RunPulse

Group
1
2
3

46.684174
47.505800
52.461667

10.776957
10.280000
.

170.739130
.
.

The Missing Data Patterns table lists distinct missing data patterns with corresponding frequencies and percents. Here, X means that the variable is observed
in the corresponding group and . means that the variable is missing. The table also
displays group-specific variable means.
Output 9.2.3.

Variance Information
The MI Procedure

Multiple Imputation Variance Information

Variable

-----------------Variance----------------Between
Within
Total

RunTime
RunPulse

0.001068
1.147555

0.059100
4.686646

0.060382
6.063711

DF
27.498
17.006

Multiple Imputation Variance Information

Variable

Relative
Increase
in Variance

Fraction
Missing
Information

RunTime
RunPulse

0.021688
0.293828

0.021448
0.246288

After the completion of m imputations, the Multiple Imputation Variance Information table displays the between-imputation variance, within-imputation variance,
and total variance for combining complete-data inferences. It also displays the degrees of freedom for the total variance. The relative increase in variance due to miss-

SAS OnlineDoc: Version 8

Example 9.2.

Propensity Score Method

183

ingness and the fraction of missing information for each variable are also displayed.
A detailed description of these statistics is provided in the Combining Inferences
from Multiply Imputed Data Sets section on page 173.
The Multiple Imputation Parameter Estimates table displays the estimated mean
and standard error of the mean for each variable. The inferences are based on the
t-distributions. For each variable, the table also displays a 95% mean confidence
interval and a t-statistic with the associated p-value for the hypothesis that the population mean is equal to the value specified in the MU0= option, which is zero by
default.
Output 9.2.4.

Parameter Estimates
The MI Procedure
Multiple Imputation Parameter Estimates

Variable

Mean

Std Error

RunTime
RunPulse

10.603677
170.400000

0.245727
2.462460

95% Confidence Limits

10.0999
165.2048

11.1074
175.5952

DF
27.498
17.006

Multiple Imputation Parameter Estimates

Variable

Minimum

Maximum

Mu0

t for H0:
Mean=Mu0

Pr > |t|

RunTime
RunPulse

10.558065
168.967742

10.648387
171.838710

0
0

43.15
69.20

<.0001
<.0001

The following statements list the first ten observations of the data set outpscore.
proc print data=outpscore(obs=10);
title First 10 Observations of the Imputed Data Set;
run;

Output 9.2.5.

Imputed Data Set

First 10 Observations of the Imputed Data Set

Obs
1
2
3
4
5
6
7
8
9
10

_Imputation_

Oxygen

Run
Time

Run
Pulse

1
1
1
1
1
1
1
1
1
1

44.609
45.313
54.297
59.571
49.874
44.811
45.681
49.091
39.442
60.055

11.37
10.07
8.65
8.63
9.22
11.63
11.95
10.85
13.08
8.63

178
185
156
146
156
176
176
156
174
170

SAS OnlineDoc: Version 8

184 

Chapter 9. The MI Procedure

Example 9.3. Regression Method

This example uses the regression method to impute missing values in a data set with
a monotone missing pattern. The following statements invoke the MI procedure and
request the regression method. The resulting data set is named outreg.
proc mi data=FitMono round=.001 .01 1
seed=55417 out=outreg;
monotone method=reg;
var Oxygen RunTime RunPulse;
run;

mu0= 50 10 150

The ROUND= option is used to round the imputed values to the same precision as
observed values. The values specified with the ROUND= option are matched with the
variables Oxygen, RunTime, and RunPulse in the order listed with the VAR statement. The MU0= option requests t tests for the hypotheses that the population means
corresponding to the variables in the VAR statement are Oxygen=50, RunTime=10,
and RunPulse=150.
The Missing Data Patterns table lists distinct missing data patterns with corresponding frequencies and percents. It is identical to the table in the previous example.
After the completion of five imputations by default, the Multiple Imputation Variance Information table displays the between-imputation variance, within-imputation
variance, and total variance for combining complete-data inferences. The relative increase in variance due to missingness and the fraction of missing information for
each variable are also displayed. These statistics are described in the Combining
Inferences from Multiply Imputed Data Sets section on page 173.
Output 9.3.1.

Variance Information
The MI Procedure

Multiple Imputation Variance Information

Variable

-----------------Variance----------------Between
Within
Total

RunTime
RunPulse

0.004443
1.790531

0.068684
4.045134

0.074016
6.193770

DF
25.294
11.846

Multiple Imputation Variance Information

Variable

Relative
Increase
in Variance

Fraction
Missing
Information

RunTime
RunPulse

0.077629
0.531166

0.074435
0.382947

The Multiple Imputation Parameter Estimates table displays a 95% mean confidence interval and a t-statistic with its associated p-value for each of the hypotheses
requested with the MU0= option.

SAS OnlineDoc: Version 8

Example 9.4.
Output 9.3.2.

MCMC Method

185

Parameter Estimates
The MI Procedure
Multiple Imputation Parameter Estimates

Variable

Mean

Std Error

RunTime
RunPulse

10.575871
170.425806

0.272059
2.488729

95% Confidence Limits

10.0159
164.9955

11.1359
175.8561

DF
25.294
11.846

Multiple Imputation Parameter Estimates

Variable

Minimum

Maximum

Mu0

t for H0:
Mean=Mu0

Pr > |t|

RunTime
RunPulse

10.506452
169.290323

10.680968
171.935484

10.000000
150.000000

2.12
8.21

0.0443
<.0001

The following statements list the first ten observations of the data set outreg. Note
that the imputed values rounded to the same precision as the observed values.
proc print data=outreg(obs=10);
title First 10 Observations of the Imputed Data Set;
run;

Output 9.3.3.

Imputed Data Set

First 10 Observations of the Imputed Data Set

Obs
1
2
3
4
5
6
7
8
9
10

_Imputation_

Oxygen

Run
Time

Run
Pulse

1
1
1
1
1
1
1
1
1
1

44.609
45.313
54.297
59.571
49.874
44.811
45.681
49.091
39.442
60.055

11.37
10.07
8.65
7.18
9.22
11.63
11.95
10.85
13.08
8.63

178
185
156
156
192
176
176
174
174
170

Example 9.4. MCMC Method

This example uses the MCMC method to impute missing values for a data set with
an arbitrary missing pattern. The following statements invoke the MI procedure and
specify the MCMC method with three imputations.
proc mi data=FitMiss seed=55417 nimpute=3 mu0=50 10 180;
mcmc chain=multiple displayinit initial=em(itprint);
var Oxygen RunTime RunPulse;
run;

SAS OnlineDoc: Version 8

186 

Chapter 9. The MI Procedure

Output 9.4.1.

Model Information
The MI Procedure
Model Information

Data Set
Method
Multiple Imputation Chain
Initial Estimates for MCMC
Start
Prior
Number of Imputations
Number of Burn-in Iterations
Seed for random number generator

WORK.FITMISS
MCMC
Multiple Chains
EM Posterior Mode
Starting Value
Jeffreys
3
200
55417

With CHAIN=MULTIPLE, the procedure uses multiple chains and completes the default 200 burn-in iterations before each imputation. The 200 burn-in iterations are
used to make the iterations converge to the stationary distribution before the imputation.
By default, the procedure uses a noninformative Jeffreys prior to derive the posterior
mode from the EM algorithm as the starting values for the MCMC process.
The following Missing Data Patterns table lists distinct missing data patterns with
corresponding statistics.
Output 9.4.2.

Missing Data Patterns

The MI Procedure
Missing Data Patterns

Group
1
2
3
4
5

Oxygen

Run
Time

Run
Pulse

X
X
X
.
.

X
X
.
X
X

X
.
.
X
.

Freq

Percent

21
4
3
1
2

67.74
12.90
9.68
3.23
6.45

Missing Data Patterns

Group
1
2
3
4
5

-----------------Group Means---------------Oxygen
RunTime
RunPulse
46.353810
47.109500
52.461667
.
.

10.809524
10.137500
.
11.950000
9.885000

171.666667
.
.
176.000000
.

With the ITPRINT option in INITIAL=EM, the procedure also displays the EM
(Posterior Mode) Iteration History table.

SAS OnlineDoc: Version 8

Example 9.4.
Output 9.4.3.

MCMC Method

187

EM (Posterior Mode) Iteration History

The MI Procedure
EM (Posterior Mode) Iteration History

_Iteration_

-2 Log L

-2 Log Posterior

Oxygen

RunTime

0
1
2
3
4
5
6
7

254.482800
255.081159
255.271405
255.318621
255.330259
255.333160
255.333896
255.334085

282.909590
282.051588
282.017488
282.015372
282.015232
282.015222
282.015222
282.015222

47.104086
47.104079
47.104077
47.104002
47.103861
47.103797
47.103774
47.103766

10.554864
10.554859
10.554858
10.554524
10.554388
10.554341
10.554325
10.554320

EM (Posterior Mode) Iteration History

_Iteration_

RunPulse

0
1
2
3
4
5
6
7

171.381796
171.381708
171.381669
171.381853
171.382058
171.382152
171.382186
171.382197

With the DISPLAYINIT option in the MCMC statement, the following Initial Parameter Estimates for MCMC table displays the starting mean and covariance estimates used in MCMC. The same starting estimates are used for the MCMC process
for multiple chains because the EM algorithm is applied to the same data set in each
chain. You can explicitly specify different initial estimates for different imputations,
or you can use the bootstrap to generate different parameter estimates from the EM
algorithm for the MCMC process.
Output 9.4.4.

Initial Parameter Estimates

The MI Procedure
Initial Parameter Estimates for MCMC

_TYPE_

_NAME_

MEAN
COV
COV
COV

Oxygen
RunTime
RunPulse

Oxygen

RunTime

RunPulse

47.103766
24.549968
-5.726112
-15.926034

10.554320
-5.726112
1.781407
3.124798

171.382197
-15.926034
3.124798
83.164044

The following two tables display variance information and parameter estimates from
the multiple imputation.

SAS OnlineDoc: Version 8

188 

Chapter 9. The MI Procedure

Output 9.4.5.

Variance Information
The MI Procedure

Multiple Imputation Variance Information

-----------------Variance----------------Between
Within
Total

Variable
Oxygen
RunTime
RunPulse

0.009200
0.002255
0.043126

0.987880
0.069112
3.650388

1.000148
0.072119
3.707889

DF
27.778
26.388
27.653

Multiple Imputation Variance Information

Variable

Relative
Increase
in Variance

Fraction
Missing
Information

Oxygen
RunTime
RunPulse

0.012418
0.043503
0.015752

0.012414
0.043351
0.015744

Output 9.4.6.

Parameter Estimates
The MI Procedure
Multiple Imputation Parameter Estimates

Variable

Mean

Std Error

Oxygen
RunTime
RunPulse

47.198228
10.510911
172.113649

1.000074
0.268549
1.925588

95% Confidence Limits

45.1489
9.9593
168.1670

49.2475
11.0625
176.0603

DF
27.778
26.388
27.653

Multiple Imputation Parameter Estimates

Variable

Minimum

Maximum

Mu0

t for H0:
Mean=Mu0

Pr > |t|

Oxygen
RunTime
RunPulse

47.132351
10.456079
171.943144

47.308274
10.538446
172.344920

50.000000
10.000000
180.000000

-2.80
1.90
-4.10

0.0092
0.0681
0.0003

Example 9.5. Producing Monotone Missingness with MCMC

This example uses the MCMC method to impute just enough missing values for a data
set with an arbitrary missing pattern so that each imputed data set has a monotone
missing pattern based on the order of variables in the VAR statement.
The following statements invoke the MI procedure and specify the the IMPUTE=MONOTONE option to create the imputed data set with a monotone missing
pattern. You must specify a VAR list to provide the order of variables for the imputed
data to achieve a monotone missing pattern.

SAS OnlineDoc: Version 8

Example 9.5.

Producing Monotone Missingness with MCMC

189

proc mi data=FitMiss seed=55417 out=outmono;

mcmc impute=monotone;
var Oxygen RunTime RunPulse;
run;

Output 9.5.1.

Model Information
The MI Procedure
Model Information

Data Set
Method
Multiple Imputation Chain
Initial Estimates for MCMC
Start
Prior
Number of Imputations
Number of Burn-in Iterations
Number of Iterations
Seed for random number generator

WORK.FITMISS
Monotone-data MCMC
Single Chain
EM Posterior Mode
Starting Value
Jeffreys
5
200
100
55417

The following Missing Data Patterns table lists distinct missing data patterns with
corresponding statistics. Here, an X means that the variable is observed in the
corresponding group, a . means that the variable is missing and will be imputed to
achieve the monotone missingness for the imputed data set, and an O means that
the variable is missing and will not be imputed. The table also displays group-specific
variable means.
Output 9.5.2.

Missing Data Pattern

The MI Procedure
Missing Data Patterns

Group
1
2
3
4
5

Oxygen

Run
Time

Run
Pulse

X
X
X
.
.

X
X
O
X
X

X
O
O
X
O

Freq

Percent

21
4
3
1
2

67.74
12.90
9.68
3.23
6.45

Missing Data Patterns

Group
1
2
3
4
5

-----------------Group Means---------------Oxygen
RunTime
RunPulse
46.353810
47.109500
52.461667
.
.

10.809524
10.137500
.
11.950000
9.885000

171.666667
.
.
176.000000
.

SAS OnlineDoc: Version 8

190 

Chapter 9. The MI Procedure

As shown in the table, the MI procedure only needs to impute three missing values
from Group 4 and Group 5 to achieve a monotone missing pattern for the imputed
data set.
When using the MCMC method to produce an imputed data set with a monotone
missing pattern, tables of variance information and parameter estimates are not created.
The following statements are used just to show the monotone missingness of the
output data set outmono.
proc mi data=outmono ( where= (_Imputation_=1) )
nimpute=0;
var Oxygen RunTime RunPulse;
run;

Output 9.5.3.

Monotone Missing Data Pattern

The MI Procedure
Missing Data Patterns

Group
1
2
3

Oxygen

Run
Time

Run
Pulse

X
X
X

X
X
.

X
.
.

Freq

Percent

22
6
3

70.97
19.35
9.68

Missing Data Patterns

Group
1
2
3

-----------------Group Means---------------Oxygen
RunTime
RunPulse
46.307744
46.372151
52.461667

10.861364
10.053333
.

171.863636
.
.

The following statements impute one value for each missing value in the monotone
missingness data set outmono. The variable Imputation is renamed to Impute
so that it will not be overwritten by the the new variable Imputation being created
in the MI procedure.
proc mi data=outmono( rename=(_Imputation_=Impute))
nimpute=1 seed=43672
out=outds( rename=(Impute=_Imputation_) drop=_Imputation_);
monotone method=reg;
var Oxygen RunTime RunPulse;
by Impute;
run;

SAS OnlineDoc: Version 8

Example 9.6.

Checking Convergence in MCMC

191

The variable Impute is renamed to Imputation in the output data outds. This
makes the output data set have the same structure as output data sets generated from
other imputation methods. You can then analyze these data sets by using other SAS
procedures and combine these results by using the procedure MIANALYZE. Note
that the VAR statement is required with a MONOTONE statement to provide the
variable order for the monotone missing pattern.

Example 9.6. Checking Convergence in MCMC

This example uses the MCMC method with a single chain. It also displays time-series
and autocorrelation plots to check convergence for the single chain.
The following statements use the MCMC method to create an iteration plot for the
successive estimates of the mean of Oxygen. Note that iterations during the burn-in
period are indicated with negative iteration numbers. These statements also create an
autocorrelation function plot for the variable Oxygen.
proc mi data=FitMiss seed=37921 noprint nimpute=2;
mcmc timeplot(mean(Oxygen)) acfplot(mean(Oxygen));
var Oxygen RunTime RunPulse;
run;
Output 9.6.1.

Time-Series Plot for Oxygen

By default, the MI procedure uses the plus sign (+) as the plot symbol to display
the points in the plot. The time-series plot shows no apparent trends for the variable
Oxygen.

SAS OnlineDoc: Version 8

192 

Chapter 9. The MI Procedure

Output 9.6.2.

Autocorrelation Function Plot for Oxygen

By default, the MI procedure uses the star sign (*) as the plot symbol to display the
points in the plot, a solid line to display the reference line of zero autocorrelation,
and a pair of dashed lines to display approximately 95% confidence limits for the
autocorrelations. The autocorrelation function plot shows no significant positive or
negative autocorrelation.
The following statements use display options to modify the autocorrelation function
plot for Oxygen.
proc mi data=FitMiss seed=37921 noprint nimpute=2;
mcmc acfplot(mean(Oxygen) / symbol=dot lref=2);
var Oxygen RunTime RunPulse;
run;

SAS OnlineDoc: Version 8

Example 9.6.
Output 9.6.3.

Checking Convergence in MCMC

193

Modified Autocorrelation Function Plot for Oxygen

You can also create plots for the worst linear function, the means of other variables,
the variances of variables, and covariances between variables. Alternatively, you can
use the OUTITER option to save statistics such as the means, standard deviations,
covariances, -2 log LR statistic, -2 log LR statistic of the posterior mode, and worst
linear function from each iteration in an output data set. Then you can do a more
in-depth time-series analysis of the iterations with other procedures, such as PROC
AUTOREG and PROC ARIMA in the SAS/ETS Users Guide, Version 8.

SAS OnlineDoc: Version 8

194 

Chapter 9. The MI Procedure

Example 9.7. Transformation to Normality

This example applies the MCMC method to the FitMiss data set in which the variable
Oxygen is transformed. Assume that Oxygen is skewed and can be transformed to
normality with a logarithmic transformation. The following statements invoke the MI
procedure and specify the transformation. The TRANSFORM statement specifies the
log transformation for Oxygen. Note that the values displayed for Oxygen in all of
the results correspond to transformed values.
proc mi data=FitMiss seed=37921 mu0=50 10 180 out=outmi;
transform log(Oxygen);
mcmc chain=multiple displayinit;
var Oxygen RunTime RunPulse;
run;

The following Missing Data Patterns table lists distinct missing data patterns with
corresponding statistics for the FitMiss data. Note that the values of Oxygen shown
in the tables are transformed values.
Output 9.7.1.

Missing Data Pattern

The MI Procedure
Missing Data Patterns

Group
1
2
3
4
5

Oxygen

Run
Time

Run
Pulse

X
X
X
.
.

X
X
.
X
X

X
.
.
X
.

Freq

Percent

21
4
3
1
2

67.74
12.90
9.68
3.23
6.45

Transformed Variables: Oxygen

Missing Data Patterns

Group
1
2
3
4
5

-----------------Group Means---------------Oxygen
RunTime
RunPulse
3.829760
3.851813
3.955298
.
.

10.809524
10.137500
.
11.950000
9.885000

Transformed Variables: Oxygen

SAS OnlineDoc: Version 8

171.666667
.
.
176.000000
.

Example 9.7.

Transformation to Normality

195

The following Variable Transformations table lists the variables that have been
transformed.
Output 9.7.2.

Missing Data Pattern

The MI Procedure
Variable Transformations
Variable

_Transform_

Oxygen

LOG

The following Initial Parameter Estimates for MCMC table displays the starting
mean and covariance estimates used in the MCMC process.
Output 9.7.3.

Initial Parameter Estimates

The MI Procedure
Initial Parameter Estimates for MCMC

_TYPE_

_NAME_

MEAN
COV
COV
COV

Oxygen
RunTime
RunPulse

Oxygen

RunTime

RunPulse

3.846122
0.010827
-0.120891
-0.328772

10.557605
-0.120891
1.744580
3.011179

171.382949
-0.328772
3.011179
82.747608

Transformed Variables: Oxygen

SAS OnlineDoc: Version 8

196 

Chapter 9. The MI Procedure

The following table displays variance information from the multiple imputation.
Output 9.7.4.

Variance Information
The MI Procedure
Multiple Imputation Variance Information
-----------------Variance----------------Between
Within
Total

Variable
* Oxygen
RunTime
RunPulse

0.000004541
0.000814
0.182700

0.000398
0.063128
3.498974

0.000404
0.064105
3.718214

DF
27.766
27.708
25.923

* Transformed Variables
Multiple Imputation Variance Information

Variable

Relative
Increase
in Variance

Fraction
Missing
Information

* Oxygen
RunTime
RunPulse

0.013685
0.015478
0.062658

0.013590
0.015356
0.060595

* Transformed Variables

The following table displays parameter estimates from the multiple imputation. Note
that the parameter value of Mu0 has also been transformed using the logarithmic
transformation.
Output 9.7.5.

Parameter Estimates
The MI Procedure
Multiple Imputation Parameter Estimates

Variable

Mean

Std Error

* Oxygen
RunTime
RunPulse

3.845991
10.586242
170.849654

0.020091
0.253190
1.928267

95% Confidence Limits

3.8048
10.0674
166.8855

3.8872
11.1051
174.8138

DF
27.766
27.708
25.923

* Transformed Variables
Multiple Imputation Parameter Estimates

Variable

Minimum

Maximum

Mu0

t for H0:
Mean=Mu0

Pr > |t|

* Oxygen
RunTime
RunPulse

3.843860
10.547440
170.315955

3.848775
10.616746
171.324638

3.912023
10.000000
180.000000

-3.29
2.32
-4.75

0.0028
0.0282
<.0001

* Transformed Variables

SAS OnlineDoc: Version 8

Example 9.7.

Transformation to Normality

197

The following statements list the first ten observations of the data set outmi. Note
that the values for Oxygen are in the original scale.
proc print data=outmi(obs=10);
title First 10 Observations of the Imputed Data Set;
run;

Output 9.7.6.

Imputed Data Set in Original Scale

First 10 Observations of the Imputed Data Set

Obs
1
2
3
4
5
6
7
8
9
10

_Imputation_

Oxygen

RunTime

Run
Pulse

1
1
1
1
1
1
1
1
1
1

44.6090
45.3130
54.2970
59.5710
49.8740
44.8110
43.4130
44.6435
39.4420
60.0550

11.3700
10.0700
8.6500
8.4840
9.2200
11.6300
11.9500
10.8500
13.0800
8.6300

178.000
185.000
156.000
155.503
166.031
176.000
176.000
173.761
174.000
170.000

The preceding results can also be produced from the following statements without
using a TRANSFORM statement.
data temp;
set FitMiss;
LogOxygen= log(Oxygen);
run;
proc mi data=temp seed=37921 mu0=3.91202 10 180 out=outtemp;
mcmc chain=multiple displayinit;
var LogOxygen RunTime RunPulse;
run;
data outmi;
set outtemp;
Oxygen= exp(LogOxygen);
run;

Note that a transformed value of log(50)=3.91202 is used in the MU0= option.

SAS OnlineDoc: Version 8

198 

Chapter 9. The MI Procedure

Example 9.8. Saving and Using Parameters for MCMC

This example uses the MCMC method with multiple chains as specified in Example 9.4. It saves the parameter values used for each imputation in an output data set
of type EST. This output data set can then be used to impute missing values in other
similar input data sets. The following statements invoke the MI procedure and specify
the MCMC method with multiple chains to create three imputations.
proc mi data=FitMiss seed=55417 nimpute=3 mu0=50 10 180 noprint;
mcmc chain=multiple outest=miest;
var Oxygen RunTime RunPulse;
run;

The following statements list the parameters used for the imputations. Note that the
data set includes observations with TYPE =SEED containing the seed to start the
next random number generator.
proc print data=miest;
title Parameters for the Imputations;
run;

Output 9.8.1.

OUTEST Data Set

Parameters for the Imputations

Obs _Imputation_ _TYPE_

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

1
1
1
1
1
2
2
2
2
2
3
3
3
3
3

SEED
PARM
COV
COV
COV
SEED
PARM
COV
COV
COV
SEED
PARM
COV
COV
COV

_NAME_

Oxygen
RunTime
RunPulse

Oxygen
RunTime
RunPulse

Oxygen
RunTime
RunPulse

Oxygen

RunTime

RunPulse

2099769086.00
49.31
32.05
-7.47
-28.32
419117425.00
47.49
41.02
-8.60
-34.29
535522494.00
45.98
43.24
-9.90
8.14

2099769086.00
10.00
-7.47
2.41
6.75
419117425.00
10.43
-8.60
2.25
7.61
535522494.00
10.82
-9.90
2.75
-2.72

2099769086.00
172.19
-28.32
6.75
128.61
419117425.00
171.58
-34.29
7.61
142.94
535522494.00
172.45
8.14
-2.72
218.32

The following statements invoke the MI procedure and use the INEST= option in the
MCMC statement.
proc mi data=FitMiss;
mcmc inest=miest;
var Oxygen RunTime RunPulse;
run;

SAS OnlineDoc: Version 8

References
Output 9.8.2.

199

Model Information
The MI Procedure
Model Information

Data Set
Method
INEST Data Set
Number of Imputations

WORK.FITMISS
MCMC
WORK.MIEST
3

The remaining tables for the example are identical to the tables in Example 9.4.

References
Anderson, T.W. (1984), An Introduction to Multivariate Statistical Analysis, Second
Edition, New York: John Wiley & Sons, Inc.
Allison, P.D. (2000), Multiple Imputation for Missing Data: A Cautionary Tale,
Sociological Methods and Research, 28, 301309.
Barnard, J. and Rubin, D.B. (1999), Small-Sample Degrees of Freedom with Multiple Imputation, Biometrika, 86, 948955.
Dempster, A.P., Laird, N.M., and Rubin, D.B. (1977), Maximum Likelihood from
Incomplete Data via the EM Algorithm, Journal of the Royal Statistical Society,
Ser. B., 39, 138.
Gelman, A. and Rubin, D.B. (1992), Inference from Iterative Simulation Using Multiple Sequences, Statistical Science, 7, 457472.
Goodnight, J.H. (1979), A Tutorial on the Sweep Operator, American Statistician,
33, 149158.
Lavori, P.W., Dawson, R., and Shera, D. (1995), A Multiple Imputation Strategy
for Clinical Trials with Truncation of Patient Data, Statistics in Medicine, 14,
19131925.
Li, K.H. (1988), Imputation Using Markov Chains, Journal of Statistical Computation and Simulation, 30, 5779.
Li, K.H., Raghunathan, T.E., and Rubin, D.B. (1991), Large-Sample Significance
Levels from Multiply Imputed Data Using Moment-Based Statistics and an F
Reference Distribution, Journal of the American Statistical Association, 86,
10651073.
Little, R.J.A. and Rubin, D.B. (1987), Statistical Analysis with Missing Data, New
York: John Wiley & Sons, Inc.
Liu, C. (1993), Bartletts Decomposition of the Posterior Distribution of the Covariance for Normal Monotone Ignorable Missing Data, Journal of Multivariate
Analysis, 46, 198206.
McLachlan, G.J. and Krishnan, T. (1997), The EM Algorithm and Extensions, New
York: John Wiley & Sons, Inc.
SAS OnlineDoc: Version 8

200 

Chapter 9. The MI Procedure

Rosenbaum, P.R. and Rubin, D.B. (1983), The Central Role of the Propensity Score
in Observational Studies for Causal Effects, Biometrika, 70, 4155.
Rubin, D.B. (1976), Inference and Missing Data, Biometrika, 63, 581592.
Rubin, D.B. (1987), Multiple Imputation for Nonresponse in Surveys, New York:
John Wiley & Sons, Inc.
Rubin, D.B. (1996), Multiple Imputation After 18+ Years, Journal of the American
Statistical Association, 91, 473489.
SAS Institute Inc. (1999), SAS/ETS Users Guide, Version 8, Cary, NC: SAS Institute
Inc.
SAS Institute Inc. (1999), SAS/GRAPH Software: Reference, Version 8, Cary, NC:
SAS Institute Inc.
SAS Institute Inc. (1999), SAS Language Reference: Concepts, Version 8, Cary, NC:
SAS Institute Inc.
SAS Institute Inc. (1999), SAS Procedures Guide, Version 8, Cary, NC: SAS Institute
Inc.
SAS Institute Inc. (1999), SAS/STAT Users Guide, Version 8, Cary, NC: SAS Institute Inc.
Schafer, J.L. (1997), Analysis of Incomplete Multivariate Data, New York: Chapman
and Hall.
Tanner, M.A. and Wong, W.H. (1987), The Calculation of Posterior Distributions
by Data Augmentation, Journal of the American Statistical Association, 82,
528540.

SAS OnlineDoc: Version 8