Macroeconomic Variables and Panel Estimation
摘要
We enhance the Bank of Italy’s statistical model for the credit assessment of non-financial firms (S-ICAS) in two ways. First, we account directly for the effects of the business cycle in the estimation of the probability of default of firms with the use of a satellite macroeconomic model. Second, to fully exploit the heterogeneity of the dataset, we experiment panel econometric methods to estimate the probability of default on a subsample of firms. The two enhancements increase the discriminatory power and the stability of S-ICAS. The macroeconomic model has already been implemented in the running system. The possible adoption of the panel econometric techniques for operational purposes requires further investigation.