Does reject inference really improve the performance of application scoring models

By Jonathan N. Crook, John Banasik. 2004

141 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Characterizing Fairness Over the Set of Good Models Under Selective Labels (2021, methodology): “One approach, Reject inference by extrapolation (RIE), uses ˆ µ ( x ) as pseudo-outcomes for the unknown observations [49].”
Inferring the outcomes of rejected loans: an application of semisupervised clustering (2019, methodology): “This is what has been found in previous literature (Banasik et al. , 2003; Chen and Astebro, 2012) and is consistent with Crook and Banasik (2004).”
A literature review on the application of evolutionary computing to credit scoring (2013, methodology): “Reject inference investigates on how to include those previous applicants in the modelling process in order to make the classifier more accurate and less biased (Crook and Banasik, 2004). This situation can become even worse with the existence of incomplete data, which may be caused by many circumst”
Deep generative models for reject inference in credit scoring (2019, background): “None of the empirical research using augmentation shows signiﬁcant improvements in either correcting the selection bias or improving model performance, see [1, 2, 26, 27, 28, 29, 31].”
Reject inference in application scorecards: evidence from France (2016, methodology): “2.3 Reweighting Reweighting is a well-used technique that involves weighting accepted applications in such a way as to scale up to the total population (Hsia, 1978; Banasik & Crook, 2004, 2007).”
Recent developments in consumer credit risk assessment (2007, background): “Experimentation has shown that the size of the bias increases monotonically if the cut-off used on the legacy scoring model increases and the accept rate decreases (Crook and Banasik, 2004).”
A rule of thumb for reject inference in credit scoring (2014, background): “A number of reject inference approaches have been developed [3] over the years.”