In SAS Model Manager, a model monitoring job runs weekly and computes PSI and KS statistics. The monitoring job configuration requires specifying a 'baseline dataset.' What role does the baseline dataset play?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — the baseline dataset is the reference distribution of predicted scores, usually from training or initial validation scoring. You use it to compare current production scores and compute PSI for drift.
Full Explanation
The baseline dataset serves as the reference distribution of predicted scores, typically from training or initial validation scoring. PSI compares current production score distributions to that baseline to quantify drift. The training-data-for-retraining claim is wrong because the baseline is for comparison, not retraining. The ground-truth-labels claim is wrong because labels support outcome metrics like KS and Gini, not PSI, which is distribution-based. The schema-and-formats claim is wrong because schema is defined in model metadata, such as fileMetadata.json.