ABSTRACT
The statistician's utterance "regression modeling involves art and science" implies a mixture of skill acquired by experience and a technique that reflects a precise application of fact or principle. This chapter examines that regression modeling involves the trilogy of art, science, and concrete poetry. As an example of the regression trilogy, the chapter makes use of a metrical "modelogue" to introduce the machine-learning technique GenIQ, an alternative to the statistical regression models. At the beginning of every day, model builders whose tasks are to predict continuous and binary outcomes are likely to be put to use the ordinary least squares (OLS) regression model and the logistic regression model (LRM), respectively. The fitness function of the OLS regression model is mean squared error (MSE), which is minimized by calculus. In 1795, Carl Friedrich Gauss at the age of 18 developed the fundamentals of least squares analysis.
