Propensity score matching and its application to risk drivers selection in financial setting
|Urszula Grzybowska , Marek J. Karwanski|
Szkoła Główna Gospodarstwa Wiejskiego (SGGW), Nowoursynowska 166, Warszawa 02-787, Poland
In credit risk scoring models are a tool used to evaluate the level of risk associated with applicants or customers. The aim of these models is not only to estimate the probability that the client will not be able to fulfil his financial commitments but also to identify and estimate the risk drivers i.e., client’s attributes that are responsible for risk occurrence.
Unfortunately, scoring models are built based on historic data stored by bank over the clients. Selection of clients is not random. This leads to systematic errors. Therefore one seeks methods that allow for a model correction that enables application of statistical inference. Quasi-experimental designs are practical solutions to this dilemma. One of such methods is Propensity Score Matching (PSM). PSM involves controlled selection of subjects ex post.The aim of our research is to apply PSM methodology to identify risk drivers in credit risk. We will present results of our research conducted on real life data. The data concern the customers signing credit agreements to finance the purchase of a car dedicated to SME (Small and Medium Enterprises). The results will be compared with scorings obtained with help of logistic regression models and data mining approach involving families of classifiers
Presentation: Oral at 8 Ogólnopolskie Sympozjum "Fizyka w Ekonomii i Naukach Społecznych", by Urszula Grzybowska
See On-line Journal of 8 Ogólnopolskie Sympozjum "Fizyka w Ekonomii i Naukach Społecznych"
Submitted: 2015-09-06 19:51 Revised: 2015-09-06 19:51