MODELLING THE METEOROLOGICAL EFFECTS ON AIR TEMPERATURE FOR KONYA CITY IN TURKEY: THE APPROACHES OF QUANTILE REGRESSION AND QUANTILE REGRESSION NEURAL NETWORKS

Abstract

In this study, we propose the use of the quantile regression and quantile regression neural networks for the relationship between monthly air temperature and various meteorological effects. Meteorological effects and air temperature differs at different points in the conditional distribution. When applied to ten years (2000-2009) of data from Konya city, results of the  quantile regression and quantile regression neural networks show that the contributions of the explanatory variables to the conditional distribution of the air temperature vary significantly. Finally, computation of conditional air temperature through both of the methods for multiple regression allows the estimation of complete density distributions that can be used for forecasting next month’s air temperature under an uncertainty framework.