The University of Southampton
University of Southampton Institutional Repository

Monitoring relationship between score and odds in a propensity scorecard

Monitoring relationship between score and odds in a propensity scorecard
Monitoring relationship between score and odds in a propensity scorecard
Propensity scorecards allow forecasting, which bank customers will soon be interested in new credits, through assessing their willingness to make application for new loans. Such scoring models are considered efficient tools to select customers for bank marketing campaigns.

Whittaker, Whitehead and Somers in their 2007 JORS paper present a new scorecard monitoring technique, which is derived from Kalman filtering. They demonstrate it for a logistic regression model estimated using the ML method, and illustrate it with an example of a dynamic mortgage scorecard. In this presentation the same technique is used but a presented approach is more general: there are assumptions neither on the scoring model specification nor on the estimation method (a common problem in practice, while using commercial software). As the estimator features are unknown, the estimator covariance is derived from the bootstrap. The focus is on a relationship between the score and the log odds for that score. That relationship is used to determine the propensity level of a customer. The log odds estimate, which represents the propensity level provided that the baseline scorecard is still up-to-date, is compared with the estimate calculated using the relationship between the updated score and the log odds. That comparison allows controlling whether the scorecard is in fact up-to-date in terms of assigning the odds. As an example a propensity scorecard is used, developed and updated based on credit bureau data.

The presented example demonstrates that a scorecard may become less up-to-date, although its performance measures such as the Gini coefficient or the KS statistic do not change considerably. Using a deteriorated scorecard may result in wrong business decisions, especially in case of a cutoff determined on the basis of a relationship between the score and the log odds for that score. Therefore, it is important to monitor that relationship carefully
Bijak, Katarzyna
5130b6b9-fbf1-44e8-9106-1dd69c6692a6
Bijak, Katarzyna
5130b6b9-fbf1-44e8-9106-1dd69c6692a6

Bijak, Katarzyna (2009) Monitoring relationship between score and odds in a propensity scorecard. Credit Scoring and Credit Control XI Conference, United Kingdom. 26 - 28 Aug 2009.

Record type: Conference or Workshop Item (Other)

Abstract

Propensity scorecards allow forecasting, which bank customers will soon be interested in new credits, through assessing their willingness to make application for new loans. Such scoring models are considered efficient tools to select customers for bank marketing campaigns.

Whittaker, Whitehead and Somers in their 2007 JORS paper present a new scorecard monitoring technique, which is derived from Kalman filtering. They demonstrate it for a logistic regression model estimated using the ML method, and illustrate it with an example of a dynamic mortgage scorecard. In this presentation the same technique is used but a presented approach is more general: there are assumptions neither on the scoring model specification nor on the estimation method (a common problem in practice, while using commercial software). As the estimator features are unknown, the estimator covariance is derived from the bootstrap. The focus is on a relationship between the score and the log odds for that score. That relationship is used to determine the propensity level of a customer. The log odds estimate, which represents the propensity level provided that the baseline scorecard is still up-to-date, is compared with the estimate calculated using the relationship between the updated score and the log odds. That comparison allows controlling whether the scorecard is in fact up-to-date in terms of assigning the odds. As an example a propensity scorecard is used, developed and updated based on credit bureau data.

The presented example demonstrates that a scorecard may become less up-to-date, although its performance measures such as the Gini coefficient or the KS statistic do not change considerably. Using a deteriorated scorecard may result in wrong business decisions, especially in case of a cutoff determined on the basis of a relationship between the score and the log odds for that score. Therefore, it is important to monitor that relationship carefully

Full text not available from this repository.

More information

Published date: 2009
Venue - Dates: Credit Scoring and Credit Control XI Conference, United Kingdom, 2009-08-26 - 2009-08-28
Organisations: Centre of Excellence for International Banking, Finance & Accounting

Identifiers

Local EPrints ID: 361325
URI: https://eprints.soton.ac.uk/id/eprint/361325
PURE UUID: 4e1a1989-e1f3-48c9-a65f-fba94a2a319c

Catalogue record

Date deposited: 23 Jan 2014 09:58
Last modified: 18 Jul 2017 03:04

Export record

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of https://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×