A modified $\chi^2$-test for uplift models with applications in marketing performance measurement
Statistics Theory
2014-01-28 v1 Statistics Theory
Abstract
Uplift, essentially being the difference between two probabilities, is a central number in marketing performance measurement. A frequent question in applications is whether the uplifts of two campaigns are significantly different. In this article we present a new -statistic which allows to answer this question by performing a statistical test. We show that this statistic is asymptotically -distributed and demonstrate its application in a real life example. By running simulations with this new and alternative approaches, we find our suggested test to exhibit a better decisive power.
Cite
@article{arxiv.1401.7001,
title = {A modified $\chi^2$-test for uplift models with applications in marketing performance measurement},
author = {Rene Michel and Igor Schnakenburg and Tobias von Martens},
journal= {arXiv preprint arXiv:1401.7001},
year = {2014}
}
Comments
18 pages, 7 figures