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 and Infographic
Here's the deal — b is correct because the baseline dataset (typically the training or initial validation scoring output) provides the reference distribution of predicted scores. PSI compares current production score distributions to this baseline to quantify distributional shift.
Full explanation below image
Full Explanation
B is correct because the baseline dataset (typically the training or initial validation scoring output) provides the reference distribution of predicted scores. PSI compares current production score distributions to this baseline to quantify distributional shift. A is wrong; the baseline is for comparison, not retraining. C is wrong; ground truth labels would be used for outcome-based metrics, not PSI (which is distribution-based). D is wrong; variable schema is defined in the model's metadata/fileMetadata.json.