The MI Procedure

 

Example 56.6 FCS Method for Continuous Variables

This example uses FCS regression methods to impute values for all continuous variables in a data set with an arbitrary missing pattern.

The following statements invoke the MI procedure and impute missing values for the Fitness1 data set:

proc mi data=Fitness1 seed=1213 nimpute=4 mu0=50 10 180 out=outex6;
   fcs nbiter=10 reg(/details);
   var Oxygen RunTime RunPulse;
run;

The NIMPUTE= option specifies the total number of imputations. The FCS statement requests multivariate imputations by FCS methods, and the NBITER=10 option (which is the default) specifies the number of burn-in iterations before each imputation.

The "Model Information"  table in Output 56.6.1 describes the method and options used in the multiple imputation process.

Output 56.6.1 Model Information
The MI Procedure

Model Information
Data Set WORK.FITNESS1
Method FCS
Number of Imputations 4
Number of Burn-in Iterations 10
Seed for random number generator 1213

The "FCS Model Specification"  table in Output 56.6.2 describes methods and imputed variables in the imputation model. With the REG option in the FCS statement, the procedure uses the regression method to impute variables RunTime, RunPulse, and Oxygen in the model.

Output 56.6.2 FCS Model Specification
FCS Model Specification
Method Imputed Variables
Regression Oxygen RunTime RunPulse

The "Missing Data Patterns" table in Output 56.6.3 lists distinct missing data patterns with corresponding frequencies and percentages. With the default ORDER=FREQ option, variables are ordered by the descending frequency counts for the missing values in the filled-in and imputation phases.

Output 56.6.3 Missing Data Patterns
Missing Data Patterns
Group Oxygen RunTime RunPulse Freq Percent Group Means
Oxygen RunTime RunPulse
1 X X X 21 67.74 46.353810 10.809524 171.666667
2 X X . 4 12.90 47.109500 10.137500 .
3 X . . 3 9.68 52.461667 . .
4 . X X 1 3.23 . 11.950000 176.000000
5 . X . 2 6.45 . 9.885000 .

When you use the DETAILS option, the parameters used in each imputation are displayed in Output 56.6.4, Output 56.6.5, and Output 56.6.6.

Output 56.6.4 FCS Regression Model for Oxygen
Regression Models for FCS Method
Imputed
Variable
Effect Imputation
1 2 3 4
Oxygen Intercept -0.000578 -0.040829 -0.100644 0.200243
Oxygen RunTime -0.706222 -0.588050 -0.732917 -0.539925
Oxygen RunPulse -0.163355 -0.211405 -0.393984 -0.156234

Output 56.6.5 FCS Regression Model for RunTime
The MI Procedure

Regression Models for FCS Method
Imputed
Variable
Effect Imputation
1 2 3 4
RunTime Intercept -0.174786 0.145997 -0.240973 -0.291107
RunTime Oxygen -0.876802 -0.630979 -0.982318 -0.879243
RunTime RunPulse -0.084348 -0.055832 -0.231270 -0.133229

Output 56.6.6 FCS Regression Model for RunPulse
The MI Procedure

Regression Models for FCS Method
Imputed
Variable
Effect Imputation
1 2 3 4
RunPulse Intercept -0.162535 -0.598755 0.078072 -0.097289
RunPulse Oxygen -0.804417 -0.544019 -0.032744 -0.335796
RunPulse RunTime -0.057307 0.215520 0.313246 0.146078

The following statements list the first 10 observations of the data set outex6 in Output 56.6.7. Note that all missing values of all variables are imputed.

proc print data=outex6(obs=10);
   title 'First 10 Observations of the Imputed Data Set';
run;

Output 56.6.7 Imputed Data Set
First 10 Observations of the Imputed Data Set

Obs _Imputation_ Oxygen RunTime RunPulse
1 1 44.6090 11.3700 178.000
2 1 45.3130 10.0700 185.000
3 1 54.2970 8.6500 156.000
4 1 59.5710 7.7722 155.233
5 1 49.8740 9.2200 153.146
6 1 44.8110 11.6300 176.000
7 1 45.3406 11.9500 176.000
8 1 36.6027 10.8500 175.250
9 1 39.4420 13.0800 174.000
10 1 60.0550 8.6300 170.000

After the completion of the specified four imputations, the "Variance Information" table in Output 56.6.8 displays the between-imputation variance, within-imputation variance, and total variance for combining complete-data inferences. The relative increase in variance due to missingness, the fraction of missing information, and the relative efficiency for each variable are also displayed. These statistics are described in the section Combining Inferences from Multiply Imputed Data Sets.

Output 56.6.8 Variance Information
Variance Information
Variable Variance DF Relative
Increase
in Variance
Fraction
Missing
Information
Relative
Efficiency
Between Within Total
Oxygen 0.078728 0.975510 1.073920 23.888 0.100880 0.096679 0.976401
RunTime 0.001464 0.071174 0.073003 27.318 0.025709 0.025473 0.993672
RunPulse 1.469522 3.666764 5.503667 11.063 0.500960 0.378278 0.913601

The "Parameter Estimates" table in Output 56.6.9 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.

Output 56.6.9 Parameter Estimates
Parameter Estimates
Variable Mean Std Error 95% Confidence Limits DF Minimum Maximum Mu0 t for H0:
Mean=Mu0
Pr > |t|
Oxygen 47.032052 1.036301 44.8927 49.1714 23.888 46.771075 47.346642 50.000000 -2.86 0.0086
RunTime 10.494632 0.270192 9.9405 11.0487 27.318 10.453740 10.544396 10.000000 1.83 0.0781
RunPulse 169.709378 2.345990 164.5495 174.8693 11.063 168.550372 170.921431 180.000000 -4.39 0.0011