Aggregation Level in Stress-Testing Models

Abstract

We explore the question of optimal aggregation level for stress-testing models when the stress test is specified in terms of aggregate macroeconomic variables but the underlying performance data are available at a loan level. We ask whether it is better to formulate models at a disaggregated level and then aggregate the predictions in order to obtain portfolio loss values or if it is better to work directly with aggregated data to forecast losses. The answer to this question depends on the data structure. Therefore, we study this question empirically, using as our laboratory a large portfolio of home equity lines of credit. All the models considered produce good in-sample fit. In out-of-sample exercises, loan-level models have large forecast errors and underpredict default probability. Average out-of-sample performance is best for county-level models. This result illustrates that aggregation level is important to consider in the loss modeling process.