An integrated genetic algorithm-conventional regression-analysis of variance for improvement of gasoline demand estimation

Abstract

This study presents an integrated algorithm for forecasting gasoline demand based on genetic algorithm (GA) with variable parameters using stochastic procedures, conventional regression and analysis of variance (ANOVA). The proposed algorithm uses ANOVA to select either GA or conventional regression for future demand estimation. It uses minimum absolute percentage of error (MAPE) when the null hypothesis in ANOVA is accepted to select from GA or regression model. The significance of the proposed algorithm is twofold. Firstly, it is flexible and identifies the best model based on the results of ANOVA and MAPE. Secondly, the proposed algorithm may identify conventional regression as the best model for future gasoline demand forecasting because of its dynamic structure, whereas previous studies assume that GA always provide the best solutions and estimation. To show the applicability and superiority of the proposed algorithm, the data for gasoline demand in Iranian agriculture sector from 1972 to 2002 is used and applied to the proposed algorithm.