Quality measures of scoring models
Abstract
One of the basic stages of constructing credit-scoring models is the assessment of their quality understood as the ability to separate reliable and unreliable borrower population. This paper focuses on methods enabling the assessment of discrimination quality, and presents the results of researches on the basis of empirical data. Apart from establishing the measure of discrimination quality, this paper refers to the issue of the assessment of the stability of results obtained by setting a confidence interval for the quality measure of the scoring model.
The full article is available to subscribers to the journal.
Author's Biography
Paweł Siarka is currently a partner in the Institute of Financial Services. He started his career in the PTF S.A., where he was acting as the Head of Credit Risk Management. Then, as the Head of the Treasury Department, he was managing the market risk in the Santander Consumer Bank in Poland. Later, as a member of the board of the Bank of European Cooperation, he was responsible for credit and market risk management.
Citation
Siarka, Paweł (2012, September 1). Quality measures of scoring models. In the Journal of Risk Management in Financial Institutions, Volume 5, Issue 4. https://doi.org/10.69554/XFSD2568.
Publications LLP