Data defines the model by dint of genetic programming, producing the best decile table.


Response-Approval Model: An Effective Approach for Implementation
Bruce Ratner, Ph.D.

Database analysts often encounter data on individuals for whom a missing profit (back-end) value is recorded as a consequence of their non-approval, although they have positive responses (front-end) to a solicitation. Among the individuals who do respond, there is a wide variation in the profit values. The literature on the estimation of a regression model based on these missing “censored” data is extensive. Ergo, the modeling of the response-approval data is in place and easy. But, the standard implementation of the model results is unreliable, as the approach is to first create a hundred-cell table, defined by crossing the ten response deciles by the ten profit deciles. Then, the analyst “cherry-picks” the best response-approval cells among the hundred cells to target. These cells are ill-chosen by virtue of cherry-picking itself. In this article, I present an easy and reliable approach to select the best response-approval cells for target marketing.

For more information about this article, call me at 516.791.3544, or e-mail, br@dmstat1.com.
My publisher owns the copyright of the article, about which this abstract addresses. The article will appear in my forthcoming book.
My publisher has granted me permission to discuss orally the article's content, but by no means provide an outline, draft or proof-ready of the article.

Sign-up for a free GenIQ webcast: Click here.