Predictive Dynamic Models for SMEs

Abstract

Considering the fundamental role played by small and medium sized enterprises (namely SMEs) in the economy of many countries and the considerable attention placed on them in the new Basel Capital Accord, we analyze a set of classical and Bayesian longitudinal models to predict SME default probability. In this contribution we present a real application based on a panel data set of German SMEs provided in our Musing European project by Creditreform, which is one of the major rating agencies for SMEs in Germany. Creditreform deals with balance sheet services, credit risk and portfolio analyses as well as consultation and support for the development of internal rating systems.