A SAS data scientist needs to compare the performance of PROC HPFOREST against PROC HPLOGISTIC using 5-fold cross-validation on the same data. Which approach correctly implements this comparison in SAS?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — you need both models tested on the same 5 folds, so create the folds with PROC SURVEYSELECT and loop through them with PROC HPFOREST and PROC HPLOGISTIC. That lets you collect validation metrics for each model on identical held-out data. The other approaches either use non-existent CV options, compare incompatible error measures, or skip the explicit SAS fold loop.
Full Explanation
The correct approach creates five explicit folds with PROC SURVEYSELECT, then runs PROC HPFOREST and PROC HPLOGISTIC in a macro loop across those folds, training on four folds and validating on the held-out fold while collecting comparable metrics, often with PROC ASSESS. This gives a controlled comparison because both models use the same fold structure and held-out data. The CV=5 MODEL-statement option is wrong because these procedures do not provide that option. Comparing HPFOREST out-of-bag error with HPLOGISTIC training fit statistics is wrong because those metrics are not comparable and do not use the same 5-fold validation. SAS Model Studio pipeline comparison is wrong here because it does not explicitly implement the requested SAS procedure-level comparison using the same 5 folds for both named procedures.