I've uncovered two issues with PLS.
1. Partial Least Squares personality in Fit Model. If Impute Missing Data enabled and Standardize X enabled, standardization calculations seem to ignore rows of data that have missing values for any X, even for those columns where values are not missing. It would seem that since impute missing data is intended to use all of the data available that the standardization calculation would use as much non-missing data as possible. Technical Support tracking number 7612032255 has an example and more details. Technical Support sent this on to the developer, but we haven't heard back yet.
2. JSL ignores "Number of Factors()" argument within the "Fit()" argument of a PLS call if "Impute Missing Data(1)" is also in the PLS call. Technical Support tracking number 7612035565 has an example.
