The NPAR1WAY Procedure

If an observation has a missing value for a response (VAR ) variable, PROC NPAR1WAY excludes the observation from the analysis of that variable. If an observation has a missing or nonpositive value for the FREQ variable, PROC NPAR1WAY excludes that observation from the analysis.

If an observation has a missing value for the CLASS variable, PROC NPAR1WAY excludes that observation from the analysis unless you specify the MISSING option in the PROC NPAR1WAY statement. If you specify the MISSING option, PROC NPAR1WAY treats missing values of the CLASS variable as a valid level.

If an observation has a missing value for a STRATA variable, PROC NPAR1WAY excludes that observation from the stratified analyses unless you specify the MISSING option in the PROC NPAR1WAY statement. If you specify the MISSING option, PROC NPAR1WAY treats missing values of the STRATA variables as valid levels.

When you specify a STRATA statement, PROC NPAR1WAY displays a "Stratum Information" table, which lists the stratum variable levels and the number of observations (frequency) in each stratum. This table summarizes the valid observations in the input data set (or BY group). An observation is valid if it has nonmissing values of the CLASS and STRATA variables (unless you specify the MISSING option) and a positive value of the FREQ variable. PROC NPAR1WAY then analyzes each response (VAR ) variable separately; each analysis includes valid observations that have nonmissing values for the individual response variable.

PROC NPAR1WAY treats missing BY variable values like any other BY variable value. The missing values form a separate, valid BY group.