TITLE: this is an example of multiple imputation for a set of
variables with missing values
DATA: FILE = ex11.5.dat;
VARIABLE: ! the following are all the variables in the data
! set:
NAMES = x1 x2 y1-y4 v1-v50 z1-z5;
! the following variables will be used to create the
! imputed data sets:
USEVARIABLES = x1 x2 y1-y4 z1-z5;
! the following variables are saved with the imputed
! data sets, but not used to create the imputed data
! sets:
AUXILIARY = v1- v10;
MISSING = ALL (999);
DATA IMPUTATION:
! the following are the variables for which missing
! data will be imputed:
IMPUTE = y1-y4 x1 (c) x2;
NDATASETS = 10;
! the following data sets will contain data for the
! variables x1 x2 y1-y4 z1-z5 v1-v10:
SAVE = missimp*.dat;
ANALYSIS: TYPE = BASIC;
OUTPUT: TECH8;