Documents disponibles écrits par cet auteur

Evaluating models for classifying customers in retail banking collections / D. J. Hand in Journal of the operational research society (JORS), Vol. 61 N° 10 (Octobre 2010) 

Public ISBD [article]  in Journal of the operational research society (JORS) > Vol. 61 N° 10 (Octobre 2010) . - pp. 1540–1547 Titre : Evaluating models for classifying customers in retail banking collections Type de document : texte imprimé Auteurs : D. J. Hand, Auteur ; F. Zhou, Auteur Année de publication : 2011 Article en page(s) : pp. 1540–1547 Note générale : Recherche opérationnelle Langues : Anglais (eng) Mots-clés : Banking  Decision  analysis  Collections  Credit  AUC  Gini  coefficient Index. décimale : 001.424 Résumé : When seeking to establish a repayment strategy with delinquent borrowers, it is useful to determine how they are likely to behave, so that an optimal use of resources can be made. We examine two behavioural classifications (‘settle immediately’ versus ‘not settle immediately’, and ‘make some repayment’ versus ‘make no repayment’) and apply a variety of rules for predicting into which class each customer is likely to belong. Since no such rule will yield perfect predictions, the way in which performance is evaluated is crucial in choosing a good rule, and hence subsequently in obtaining accurate predictions of likely future behaviour. We examine some popular standard performance evaluation criteria, showing that they have major weaknesses. We describe and illustrate the use of an alternative measure that overcomes these weaknesses. DEWEY : 001.424 ISSN : 0361-5682 En ligne : http://www.palgrave-journals.com/jors/journal/v61/n10/abs/jors2009129a.html [article] Evaluating models for classifying customers in retail banking collections [texte imprimé] / D. J. Hand, Auteur ; F. Zhou, Auteur . - 2011 . - pp. 1540–1547. Recherche opérationnelle Langues : Anglais (eng) in Journal of the operational research society (JORS) > Vol. 61 N° 10 (Octobre 2010) . - pp. 1540–1547 Mots-clés : Banking  Decision  analysis  Collections  Credit  AUC  Gini  coefficient Index. décimale : 001.424 Résumé : When seeking to establish a repayment strategy with delinquent borrowers, it is useful to determine how they are likely to behave, so that an optimal use of resources can be made. We examine two behavioural classifications (‘settle immediately’ versus ‘not settle immediately’, and ‘make some repayment’ versus ‘make no repayment’) and apply a variety of rules for predicting into which class each customer is likely to belong. Since no such rule will yield perfect predictions, the way in which performance is evaluated is crucial in choosing a good rule, and hence subsequently in obtaining accurate predictions of likely future behaviour. We examine some popular standard performance evaluation criteria, showing that they have major weaknesses. We describe and illustrate the use of an alternative measure that overcomes these weaknesses. DEWEY : 001.424 ISSN : 0361-5682 En ligne : http://www.palgrave-journals.com/jors/journal/v61/n10/abs/jors2009129a.html

Likelihood-ratio changepoint features for consumer-behaviour models / A. R. Brentnall in Journal of the operational research society (JORS), Vol. 61 N° 3 (Mars 2010) 

Public ISBD [article]  in Journal of the operational research society (JORS) > Vol. 61 N° 3 (Mars 2010) . - pp. 462–472 Titre : Likelihood-ratio changepoint features for consumer-behaviour models Type de document : texte imprimé Auteurs : A. R. Brentnall, Auteur ; M. J. Crowder, Auteur ; D. J. Hand, Auteur Année de publication : 2011 Article en page(s) : pp. 462–472 Note générale : Recherche opérationnelle Langues : Anglais (eng) Mots-clés : Automated  teller  machine  Consumer  Finance  Prediction  Risk  Statistics Index. décimale : 001.424 Résumé : Some predictive models for customer value management might benefit from information about certain changes in individual-consumer behaviour. We take changepoint methods as the first step in producing a model-input feature for this purpose. An unusual feature in the application of changepoint methods to consumer data is there are as many streams of data as there are customers. This property is used to help decide whether an individual has changed their behaviour by ordering likelihood-ratio statistics from the changepoint models. Following a review of changepoint methods, the approach is demonstrated on cash machine transactions. Models for the amount, location and time of transaction are used and accounts exhibiting large evidence of change are examined in detail. For the data set used the approach performs sensibly. The worth of likelihood-ratio statistics to rank evidence for change is considered more generally through some of the literature. DEWEY : 001.424 ISSN : 0160-5682 En ligne : http://www.palgrave-journals.com/jors/journal/v61/n3/abs/jors2009160a.html [article] Likelihood-ratio changepoint features for consumer-behaviour models [texte imprimé] / A. R. Brentnall, Auteur ; M. J. Crowder, Auteur ; D. J. Hand, Auteur . - 2011 . - pp. 462–472. Recherche opérationnelle Langues : Anglais (eng) in Journal of the operational research society (JORS) > Vol. 61 N° 3 (Mars 2010) . - pp. 462–472 Mots-clés : Automated  teller  machine  Consumer  Finance  Prediction  Risk  Statistics Index. décimale : 001.424 Résumé : Some predictive models for customer value management might benefit from information about certain changes in individual-consumer behaviour. We take changepoint methods as the first step in producing a model-input feature for this purpose. An unusual feature in the application of changepoint methods to consumer data is there are as many streams of data as there are customers. This property is used to help decide whether an individual has changed their behaviour by ordering likelihood-ratio statistics from the changepoint models. Following a review of changepoint methods, the approach is demonstrated on cash machine transactions. Models for the amount, location and time of transaction are used and accounts exhibiting large evidence of change are examined in detail. For the data set used the approach performs sensibly. The worth of likelihood-ratio statistics to rank evidence for change is considered more generally through some of the literature. DEWEY : 001.424 ISSN : 0160-5682 En ligne : http://www.palgrave-journals.com/jors/journal/v61/n3/abs/jors2009160a.html