PHAB scores: proportional hazards analysis behavioural scores
- PDF / 238,516 Bytes
- 10 Pages / 595 x 794 pts Page_size
- 47 Downloads / 215 Views
#2001 Operational Research Society Ltd. All rights reserved. 0160-5682/01 $15.00 www.palgrave-journals.com/jors
PHAB scores: proportional hazards analysis behavioural scores M Stepanova and LC Thomas* University of Southampton, Southampton, UK Credit scoring is one of the most widely used applications of quantitative analysis in business. Behavioural scoring is a type of credit scoring that is performed on existing customers to assist lenders in decisions like increasing the balance or promoting new products. This paper shows how using survival analysis tools from reliability and maintenance modelling, specifically Cox’s proportional hazards regression, allows one to build behavioural scoring models. Their performance is compared with that of logistic regression. Also the advantages of using survival analysis techniques in building scorecards are illustrated by estimating the expected profit from personal loans. This cannot be done using the existing risk behavioural systems. Keywords: behavioural scoring; proportional hazards regression; risk; banking
Introduction Once an application has been approved the lender is interested in monitoring the performance and behaviour of a customer. A behaviour score is constructed using information about the customer’s activity, payments, purchases, etc, which are normally recorded monthly. The lender may wish to predict future payment amounts or purchase frequency based on the most recent behaviour score. Hence account management strategies, such as marketing campaigns or changing credit limit, are driven by behaviour score. At present, techniques similar to those applied for application scoring are used for constructing behavioural scoring systems. Logistic regression is the most common one, and as well as application characteristics used in application scoring, performance variables are also used to construct the score. It has been shown that survival analysis can be applied to estimate the time to default or to early repayment at the time of application only.1–3 This paper develops techniques based on Cox’s proportional hazards model incorporating behavioural data, such as information about monthly balance and delinquency, to develop survival analysis approaches to behavioural scoring. The definition of default used is 3 or more months delinquent. Firstly we outline the proposed technique for building behaviour scoring models in terms of proportional hazards formulae. Specifics of the actual model building process for a proportional hazards analysis behaviour score (PHAB score) are then explained. The next
*Correspondence: LC Thomas, School of Management, University of Southampton, Southampton, SO17 1BJ. E-mail: [email protected]
section presents an example of building such models on a behavioural data set from a major UK financial institution. The last two sections compare the results of this type of behavioural score with ones built using logistic regression and show how the extra information in a PHAB score can be used to estimate expected profit in the loan.
Propor
Data Loading...
