Well-performing default predictions show good discrimination and calibration. Discrimination is the ability to separate defaulters from nondefaulters. Calibration is the ability to make unbiased forecasts. I derive novel discrimination and calibration statistics to verify forecasts expressed in terms of probability under dependent observations. The test statistics’ asymptotic distributions can be derived in analytic form. Not accounting for cross correlation can result in the rejection of actually well-performing predictions, as shown in an empirical application. I demonstrate that forecasting errors must be serially uncorrelated. As a consequence, my multiperiod tests are statistically consistent.

We present a prediction model to forecast corporate defaults. In a theoretical model, under incomplete information in a market with publicly traded equity, we show that our approach must outperform ratings, Altman’s Z -score, and Merton’s distance to default. We reconcile the statistical and structural approaches under a common framework; that is, our approach nests Altman’s and Merton’s approaches as special cases. Empirically, the combined approach is indeed the most powerful predictor, and the numbers of observed defaults align well with the estimated probabilities. With a new transformation method, we obtain cycle-adjusted forecasts that still outperform ratings.