Automatic Imputation of Missing Values

Now you are ready to begin imputing values; we'll start with a run with automatic settings, but before requesting imputations, we'll set the random seed. Setting the random seed allows you to replicate the analysis exactly.

To set the random seed, from the menus choose:
Transform > Random Number Generators...

Figure 1. Random Number Generators dialog box
Select Set Active Generator.
Select Mersenne Twister.
Select Set Starting Point.
Select Fixed Value, and type 20070525 as the value.
Click OK.
To multiply impute missing data values, from the menus choose:
Analyze > Multiple Imputation > Impute Missing Data Values...

Figure 2. Impute Missing Data Values dialog
Select Months with service [tenure] through Number of people in household [reside] as variables in the imputation model.
Type telcoImputed as the dataset to which imputed data should be saved.
Click the Output tab.
Figure 3. Output tab
Select Descriptive statistics for variables with imputed values.
Click OK.