Limited Data for Forecasting a Fact of Life for Credit Risk Managers

As the pace of business accelerates, financial institutions struggle to find enough relevant data to forecast risk-related parameters for reserve calculation of loan portfolios. Short business cycles make it difficult to come up with a good forecasting methodology, even as the way that data is collected, and assumptions about that data, continue to change in the current dynamic environment. This paper seeks to address these issues as the authors propose forecasting methodologies that are efficient in when the data size is small.