The WEIGHT statement specifies numeric variables in the DATA= data set that contain weights for each similarity or dissimilarity measurement. These weights are used to compute weighted least-squares estimates. The number of WEIGHT variables must be the same as the number of VAR variables, and the variables in the WEIGHT statement must be in the same order as the corresponding variables in the VAR statement.

If the WEIGHT statement is omitted, all data within a partition are assigned equal weights.

Data with zero or negative weights are ignored in fitting the model but are included in the OUTRES= data set and in monotone transformations.