• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

News

Report by the Laboratory staff members at the XX April International Academic Conference On Economic and Social Development of the National Research University Higher School of Economics

On April 9, the Laboratory staff members Lapshin V.A. and Markov A.A. made a report "Confidence-Based Estimation of the Transient Probability Matrix in the Presence of Insufficient Data" at the XX April International Academic Conference On Economic and Social Development of the National Research University Higher School of Economics.

Abstract

The problem of data scarcity is one of the key challenges faced by the analyst in conducting research. The issue of data limitations is even more relevant for risk management due to a significant number of requirements to models by the regulator and various standards (including IFRS). We use various methods to assess the impact of data limitations on estimates of probability of default and expected credit losses, as well as to determine in which cases it is possible to increase the effectiveness of parameter estimates. We also present an analysis of a transitional probability matrix method and the impact of the amount of the data on the confidence estimates obtained via this method. We propose a methodology to determine the optimal grouping of borrowers' ratings in order to improve the quality of PD estimation. The study may be of interest to specialists in credit risks and risk management in general.
The problem of data scarcity is one of the key challenges faced by the analyst in conducting research. The issue of data limitations is even more relevant for risk management due to a significant number of requirements to models by the regulator and various standards (including IFRS). We use various methods to assess the impact of data limitations on estimates of probability of default and expected credit losses, as well as to determine in which cases it is possible to increase the effectiveness of parameter estimates. We also present an analysis of a transitional probability matrix method and the impact of the amount of the data on the confidence estimates obtained via this method. We propose a methodology to determine the optimal grouping of borrowers' ratings in order to improve the quality of PD estimation. The study may be of interest to specialists in credit risks and risk management in general.
 
Translated with www.DeepL.com/Translator (free version)