Using a variety of techniques, data analysts in database marketing aim to build models that maximize expected response and profit from solicitations. Standard techniques include the statistical methods of classical discriminant analysis, as well as logistic and ordinary regression. A recent addition in the data analysis arsenal is the GenIQ Model© - a hybrid machine learning- statistics method - which is presented in full detail below.

First, a background on the concept of optimization will be helpful since optimization techniques provide the estimation of all models. Genetic modeling is the "engine" for the GenIQ Model, and is discussed next as a machine learning optimization approach. Since the objective of database models is decile analysis optimization, i.e., to maximize expected response or profit from solicitations, I will demonstrate how the GenIQ Model serves to meet that objective. Actual case studies are presented to further explicate the potential of the GenIQ Model. For a preview of the 9-step modeling process of GenIQ, click here.  For FAQs about GenIQ, click here.

Related Articles:

1. Decile Analysis Primer: Cum Lift for Response Model 
2. “Dumb” Decile Analysis versus “Smart” Decile Analysis: Identifying Extreme Response Segments