Much of Model Risk Does Not Come from Any Model

Abstract

The use of financial models leads to model risk. Some of that risk comes from places in the “model ecosystem” other than the individual model itself. This article points out a few of those extra sources of model risk. Because financial models are always used as part of some financial business, even when each model has been validated successfully, there remain model risks coming from interactions with other parts of the business environment. The article discusses ecosystem risk, governance risk, data source risk, data integrity risk, new version risk, suitability risk, jurisdictional regulatory risk, assumption and override risk, multiple version risk, calibration risk, and model/nonmodel classification risk, with examples of each.

TOPICS: Quantitative methods, risk management