Logit and probit model within estimation of US banks' PD

Petr Gurny*, Martin Gurny

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

Abstract

In this paper a possibility of determination of financial institution's PD according to credit scoring models is discussed. First we will briefly introduce the two categories of credit scoring models logit model and probit model. In the main part of the paper we will work with the sample of almost three hundreds of commercial US banks and we will stepwise apply the above mentioned models on this sample to derive two models for estimation of PD. The goal of the paper is estimation of the PD of chosen US banks by means of the logit and probit model, discussion of the drawbacks and advantages of these models and comparison of the reached results.

Original languageEnglish
Title of host publicationProceedings of 47th EWGFM meeting
EditorsT. Tichy, M. Kopa
Place of PublicationOstrava, CZ
PublisherVSB-TECH UNIV OSTRAVA
Pages73-80
Number of pages8
Publication statusPublished - 2010
Externally publishedYes
Event47th Meeting of the EURO-Working-Group-on-Financial-Modelling (EWGFM) - Prague, Czech Republic
Duration: 28 Oct 201030 Oct 2010

Conference

Conference47th Meeting of the EURO-Working-Group-on-Financial-Modelling (EWGFM)
CountryCzech Republic
CityPrague
Period28/10/1030/10/10

Keywords

  • Probability of default
  • logit model
  • probit model

Cite this