The SURVEYIMPUTE Procedure
Displayed Output
If you use the NOPRINT option in the PROC SURVEYIMPUTE statement, the procedure does not display any output. Otherwise, PROC SURVEYIMPUTE displays results of the imputation in a collection of tables, which are described in the following subsections.
Class Level Information
If you use a CLASS statement, PROC SURVEYIMPUTE displays a "Class Level Information" table, which lists the categories of every CLASS variable that is used in the imputation. The ODS name of the "Class Level Information" table is ClassLevelInfo.
Convergence Status
If you specify METHOD=FEFI , PROC SURVEYIMPUTE displays a "Convergence Status" table, which shows the convergence status of the EM optimization routine. If the optimization routine converges, then the Status is set to 0; otherwise, the Status is set to 1. The ODS name of the "Convergence Status" table is ConvergenceStatus.
Design Summary
If you use a STRATA or CLUSTER statement, PROC SURVEYIMPUTE displays a "Design Summary" table, which provides information about the sample design. The table displays the total number of strata that are read and used, and the total number of clusters that are read and used. The ODS name of the "Design Summary" table is DesignSummary.
Hadamard Matrix
If you specify the PRINTH method-option for VARMETHOD=BRR , PROC SURVEYIMPUTE displays the Hadamard matrix that is used to construct replicates for BRR variance estimation. If you provide a Hadamard matrix by using the HADAMARD= method-option for VARMETHOD=BRR but the procedure does not use the entire matrix, the procedure displays only the rows and columns that are actually used to construct replicates. The ODS name of the "Hadamard Matrix" table is Hadamard.
Imputation Information
By default, PROC SURVEYIMPUTE displays an "Imputation Information" table, which provides information about the imputation method. The table displays the two-level name of the input data set, the name and label of the WEIGHT variable, the name and label of each STRATA variable, the name and label of each CLUSTER variable, the name of the imputation method used, and the random number seed. The ODS name of the "Imputation Information" table is ImputationInfo.
Imputation Summary
By default, PROC SURVEYIMPUTE displays an "Imputation Summary" table, which provides summary information about the imputation. The table displays the number of observations and the sum of weights for the following:
-
Nonmissing observations – all variables specified in the VAR statement have nonmissing values
-
Missing – at least one variable specified in the VAR statement has a missing value
-
Missing, Imputed – all missing values have been imputed
-
Missing, Not Imputed – no missing values are imputed
-
Missing, Partially Imputed – missing values in some variables are imputed, but missing values in some other variables are not imputed
The ODS name of the "Imputation Summary" table is ImputationSummary.
Iteration History
If you specify METHOD=FEFI , PROC SURVEYIMPUTE displays an "Iteration History" table, which provides information about the iteration history for the EM algorithm. The table displays iteration numbers, maximum absolute differences, and maximum relative absolute differences for the fractional weights over all the observations. The ODS name of the "Iteration History" table is IterationHistory.
Missing Data Patterns
By default, PROC SURVEYIMPUTE displays a "Missing Data Patterns" table, which provides information about the missing data patterns. The table displays the missing data pattern groups, "X" if the variable is observed in the group, and "." if the variable is missing in that group. In addition, it displays observation frequencies, the sum of weights, unweighted percentages, and weighted percentages for each group. The ODS name of the "Missing Data Patterns" table is MissPattern.