• Products
  • Statistics and Data Mining Solutions
  • Statistics and Data Mining Services
  • Statistics and Data Mining Resources
  • Support
  • Events
  • Company
News & Events
Home / News & Events / Credit Risk Management in Retail Banking

Credit Risk Management in Retail Banking

Date: March 25, 2004 at 8:30AM Pacific Time

Speakers:

Dr. Dirk Ocker
Head of Quantitative Research
Risk Controlling
Swiss Union of Raiffeisen Banks
Dr. Jan Beran
Professor of Statistics
Department of Mathematics and Statistics
University of Konstanz, Germany

Listen to the archived webcast.

Downloads:

Abstract
In recent years, several methodologies for measuring probability of default and portfolio credit risk have been introduced that demonstrate the benefits of using statistical models to model credit risk. However, most of the applications of these models have been focussed on portfolios of bonds, OTC products or corporate loans. The measurement of credit risk in retail portfolios has not received as much attention. Due to a number of specific features of retail markets, simple downsizing and adjustment of commercial credit risk engines is often not possible or may be unreliable.

This webcast gives a brief overview of some of these special features and provides the
corresponding statistical guidelines for risk assessment of retail portfolios, including:

  • Estimation of probability of default
  • The meaning of default dependencies
  • Modelling of aggregated loss distributions

Presenter Information
Dr. Jan Beran is Professor of Statistics at the Department of Mathematics and Statistics, University of Konstanz, Germany. He has published more than 50 papers and two books in mathematical statistics and statistical applications in engineering, medicine, environmental sciences, arts and finance.

Dr. Dirk Ocker is Head of Quantitative Research at the Swiss Union of Raiffeisen Banks. He has more than 5 years experience in credit risk modelling and is author of several articles in statistics and statistical applications in finance.