Model Selection in Functional Networks via Genetic Algorithms
Rosa Eva Pruneda, Beatriz Lacruz · 2007
Abstract—Several statistical tools and most recently Functional Networks (FN) have been used to solve nonlinear regression problems. One of the tasks associated with all of these methodologies consists of discovering the functional form of the contribution of the explanatory variables to the response variable. In this paper, we tackle this problem using functional network models (FNs). Since these models usually involve from a moderate to high number of parameters, a genetic algorithm (GA) for model selection is proposed. After an introduction of FNs and GAs, the performance of the proposed methodology is assessed using a simulation study as well as a real-life data set.