Comparing predictive ability in presence of instability over a very short time

We consider forecast comparison in the presence of instability when this affects only a short period of time. We demonstrate that global tests do not perform well in this case, as they were not designed to capture very short-lived instabilities, and their power vanishes altogether when the magnitude of the shock is very large. We then propose and discuss approaches that are more suitable to detect such situations, such as nonparametric methods like the S test from Andrews (2003) or the MAX procedure from Harvey et al. (2021). We illustrate these results in a Monte Carlo exercise and in a comparison of the nowcast of the quarterly US nominal GDP from the Survey of Professional Forecasters (SPF) against a naive benchmark of no growth, over a period that includes the GDP instability brought by the COVID-19 crisis. We recommend that the forecaster does not pool the sample, but excludes the short periods of high local instability from the evaluation exercise.