Validation of Net Models: Measuring Stability and Discriminatory Power

René Michel, Igor Schnakenburg and Tobias von Martens
Additional contact information
René Michel: Deutsche Bank AG
Igor Schnakenburg: DeTeCon International GmbH
Tobias von Martens: Deutsche Bank AG

Chapter Chapter 4 in Targeting Uplift, 2019, pp 101-120 from Springer

Abstract: Abstract Measuring the quality of scoring models is a mandatory and crucial step in the data mining process. This chapter suggests key performance indicators of model quality that have been transferred from classical (gross) scoring or were specifically designed for net scoring. For model stability, the average squared deviation, a significance-based measure, and the model stability rank correlation are regarded as appropriate, whereas for discriminatory power, Qini, AUnROC, and a significance-based measure should be used. Subsequently, the validation and adjustment of uplift models over time are explained.

Date: 2019
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-030-22625-1_4

Ordering information: This item can be ordered from
http://www.springer.com/9783030226251

DOI: 10.1007/978-3-030-22625-1_4

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().