Model-Based Estimation of Sovereign Default Risk

Abstract

We estimate a canonical sovereign default model from Arellano (2008) for Argentina via maximum simulated likelihood estimation to understand how well it performs in terms of predicting default events. The estimated model accounts for the overall default patterns of Argentina and closely matches the default data. Out-of-sample forecasting shows that the model performs better than a logit model in predicting the onset of default events. In terms of the business cycle statistics, the findings of the model are consistent with the data and Arellano (2008), with some caveats.