Resource title

Credit scoring: Discussion of methods and a case study

Resource image

image for OpenScout resource :: Credit scoring: Discussion of methods and a case study

Resource description

The scenario considered is that of a credit association, a bank or an-other nancial institution which, on the basis of information about anew potential customer and historical data on many other customers,has to decide whether or not to give that customer a certain loan.We discuss three popular techniques: logistic regression, discriminantanalysis and neural networks. We shall argue strongly in favour ofthe logistic regression. Discriminant analysis can be used, and forreasons that can be explained mathematically it will often result inapproximately the same conclusions as a logistic regression. But thestatistical assumptions are not appropriate in most cases, and theresults given are not as directly interpretable as those of logistic re-gression. Neural network techniques, in their simplest form, su erfrom the lack of statistical standard methods for veri cation of themodel and tests for removal of covariates. This problem disappearsto some extend when the neural networks are reformulated as properstatistical models, based on the type of functions that are consideredin neural networks. But this results in a somewhat specialized class ofnon{linear regression models, which may be useful in situations wherelocal peculiarities of the response function are in focus, but certainlynot when the overall | usually monotone | e ect of many more orless confounded covariates is the issue. We discuss, within the logisticregression framework, the handling of phenomena such as time trendsand corruption of the historical data due to shifts of policy, censor-ing and/or interventions in highrisk customers' economy. Finally, weillustrate and support the theoretical considerations by a case studyconcerning mortgage loans in a Danish credit associatio

Resource author

Dorte Kronborg, Tue Tjur

Resource publisher

Resource publish date

Resource language


Resource content type


Resource resource URL

Resource license

Check the according license before adaptation. When adapting give credits to the original author.