Stress Testing and the Quantification of the Dependency Structure Amongst Portfolio Segments

Abstract

A critical question that banking supervisors are trying to answer is what is the amount of capital or liquidity resources required by an institution in order to support the risks taken in the course of business. The financial crisis of the last several years has revealed that traditional approaches such as regulatory capital ratios to be inadequate, giving rise to supervisory stress testing as a primary tool. In the case of banks that model the risk of their portfolios using top-of-the-house modeling techniques, an issue that is rarely addressed is how to incorporate the correlation of risks amongst the different segments. An approach to incorporate this consideration of de-pendency structure is proposed, and the bias that results from ignoring this aspect is quantified, through estimating a vector autoregressive (“VAR”) time series models for credit loss using Fed Y9 data. We find that the multiple equation VAR model outperforms the single equation auto-regressive (“AR”) models according to various metrics across all modeling segments.

Keywords: Stress Testing, Correlation, CCAR, DFAST, Credit Risk, Financial Crisis, Model Risk, Vector Autoregression

JEL Classification: G11, G21, G22, G28, G32