PSDATA Statement
The PSDATA statement specifies the treatment indicator variable and a variable for either the propensity score or the logit of propensity score for variables that are in the DATA= data set. Either the PSMODEL statement or the PSDATA statement is required.
You must specify the following option:
-
TREATVAR=treatvar <(TREATED='level' )>
names the treatment indicator variable, treatvar, which must be a binary classification variable that is specified in the CLASS statement. The TREATED='level' suboption indicates the level that corresponds to treatment. If the TREATED='level' suboption is not specified, the first ordered level based on the formatted values is used to derive the propensity scores.
You must also specify one (and only one) of the following options:
-
PS=name
names the variable that contains propensity scores, where the variable name must be a variable in the DATA= data set.
-
LPS=name
names the variable that contains logits of propensity scores, where the variable name must be a variable in the DATA= data set.