In SAS Model Manager, a data scientist runs a monitoring report and observes that the model's KS statistic dropped from 0.55 (baseline) to 0.32 (current month). The input variable PSI values are all below 0.10. What type of drift does this pattern most likely indicate?
Select an answer to reveal the explanation.
Short Explanation
You're seeing concept drift because the model's KS dropped sharply while all input PSI values stayed below 0.10. That means the inputs didn't move much, but the relationship between inputs and target changed.
Full Explanation
The concept-drift answer is correct because a significant KS decline, from 0.55 to 0.32, with all input PSI values below 0.10, is the hallmark of concept drift: the relationship between predictors and outcome changed while input distributions remained stable. For example, economic conditions may have changed how credit utilization relates to default probability. The data-drift option is wrong because PSI below 0.10 means input distributions are stable, not substantially changed. The model-decay-due-to-software-version-changes option is wrong because a scoring-engine version change is a deployment issue, not the most likely explanation for stable inputs plus a large KS drop without direct evidence. The sampling-bias option is wrong because sampling bias would be an artifact of the monitoring process, but stable input PSI plus a large KS decline is a direct performance signal unless the monitoring dataset is non-representative or mislabeled.