Econometric Pitfalls in Stress Testing

Iftekhar Hasan, David Lynch and Akhtar Siddique

The use of econometric methods for stress testing has created significant challenges for econometricians. These challenges include issues such as multicollinearity, specification errors, identification issues and regime dependence, that get altered by stress testing. Additionally, the short data span and different objectives pose new problems. However, they have also inspired econometricians to develop innovative methodologies. This chapter reviews several elements of these challenges as well as ways that have been proposed to overcome them.

WHY STRESS TESTS POSE CHALLENGES FOR ECONOMETRICIANS

There are several reasons why stress tests pose challenges for traditional econometric methods. First, many of the problems that occur in traditional econometric applications recur in stress testing applications, often with a twist. These are multicollinearity, specification errors such as omitted variables bias, identification issues and regime dependence. Additional complications commonly arise from the application of the econometric models to stress-test applications where the independent variables are stressed.

The second set of issues arises from the characteristics of the data

To continue reading...

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here: