Commercial Bank Credit Risk Forecasting

About the author:

Tatуana A. Zelenina,

Abstract:

In the article are presented the results of forecasting the risk of the customer loan portfolio of commercial bank on the basis of the binary choice model. As an indication of the credit risk is considered outstanding share of the portfolio of loans, as a measure of credit risk — the probability of increasing the share of outstanding of loan. Forecasting stock indexes that influence the probability of increasing the share of outstanding of loan, carried out on the basis of singular spectrum analysis and adaptive models with damped trend. Evaluation of the accuracy of models forecasting stock index performed using historical forecasts and the mean absolute percentage error of the forecast. The study shows that for 2012 is expected to decrease the probability amplitude fluctuations increase in the share of outstanding of loan, and by the middle of the third quarter and year-end is expected to increase the risk for the client portfolio.

References: