Title:A Multi-objective Procedure for Efficient Regression Modeling

Abstract:Variable selection is recognized as one of the most critical steps in statistical modeling. The problems encountered in engineering and social sciences are commonly characterized by over-abundance of explanatory variables, non-linearities and unknown interdependencies between the regressors. An added difficulty is that the analysts may have little or no prior knowledge on the relative importance of the variables. To provide a robust method for model selection, this paper introduces a technique called the Multi-objective Genetic Algorithm for Variable Selection (MOGA-VS) which provides the user with an efficient set of regression models for a given data-set. The algorithm considers the regression problem as a two objective task, where the purpose is to choose those models over the other which have less number of regression coefficients and better goodness of fit. In MOGA-VS, the model selection procedure is implemented in two steps. First, we generate the frontier of all efficient or non-dominated regression models by eliminating the inefficient or dominated models without any user intervention. Second, a decision making process is executed which allows the user to choose the most preferred model using visualizations and simple metrics.