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

Building Statistical Regression Models: Straight Data are Necessary
Bruce Ratner, Ph.D.

The quotidian techniques for predicting a continuous and binary target variable are the statistical ordinary least squares (OLS), and logistic regression models, respectively. These regression methods are called linear models because the target variable (Y) is expressed as a “linear combination” of the regression coefficients (bi), that is, as a weighted sum of the predictor variables (X1, X2, …, Xn ):

Y = b0 + b1X1 + b2X2 +… + bnXn , where the weights are the regression coefficients bi, and X0=1.

The linearity is an assumption that implies a necessary condition for building statistical regression models: Each predictor variable has a linear or straight-line relationship with the target variable. A popular technique for straightening data is Tukey’s Bulging Rule (TBR). However, if the predictor-target variable relationship has a “kink,” the TBR will not work. Notwithstanding kinks, TBR must be applied to each and every predictor variable, one at a time against the target variable. TBR is very time-consuming. The purpose of this article is to present, via a simply illustration, the GenIQ Model© as a data-straightner, which is robust, powerful, and without restrictions or assumptions. Moreover, GenIQ can be applied to many variables – at one time. GenIQ is automatic, and thusly a time-saver. For the typification of GenIQ as a data-straightener, please click here.

I would greatly appreciate your comments about GenIQ as a data-straightener. Please email me. Thank you. Bruce

For another excellent illustration where GenIQ serves as a "data straightener" - seeking the maximum predictive power of a variable as well as providing the necessary (but not sufficient) any-model assumption of the relationship between target and predictive variables is "straight," go here.

For more information about this article, call Bruce Ratner at 516.791.3544 or 1 800 DM STAT-1; or e-mail at
Sign-up for a free GenIQ webcast: Click here.