Tail Model Risk in a Stress Scenario for House Prices

This paper incorporates uncertainties of model risk in a stress scenario for house prices. Our approach consists of mapping the Gaussian (or other alternative) distribution quantiles to the quantiles of the empirical distribution using a statistical criterion. The mapping corrects for the presence of fatter tails (relative to the Gaussian distribution) into the re-parameterized distribution. The focus of the analysis is at the tail of the distribution, which is associated with uncertain adverse events. The application of the model to house prices indicates the relevance of our mapping in terms of capturing the home price declines observed in the Great Recession. The derivation of confidence intervals, furthermore, provides buffers for model risk. For example, while regulatory capital may use the point estimate of the quantile of the distribution, risk management organizations may use this confidence intervals to bolster resilience of financial institutions to home price declines