Bayesian Model Averaging (BMA) provides a coherent mechanism to address the problem of model uncertainty. In this paper we extend the BMA framework to panel data models where the lagged dependent variable as well as endogenous variables appear as regressors. We propose a Limited Information...

Intro -- Contents -- I. INTRODUCTION -- II. THE BASIC MODEL SETTING -- III. MODEL 1: A SIMPLE MODEL WITH NON-RANDOM DEFAULT PROBABILITIES -- IV. INTRODUCING THE POISSON APPROXIMATION -- V. MODEL 2: THE MODEL WITH KNOWN PROBABILITIES REVISITED -- VI. MODEL 3: THE MODEL WITH RANDOM DEFAULT...