Application of the non-linear EGARCH model in the modeling of the evapotranspiration values

In multivariate models, the modeling and predicting various parameters can improve by involving other factors. Also Since nonlinear models with conditional variance, the remaining portion of the linear models to adequately model, we expect that the combination of linear and nonlinear models, partly to increase the accuracy of modeling and predictions. In this study, were used the potential evapotranspiration values of stations in the provinces (Birjand, Mashhad, Zahedan and Zabol stations) during the statistical period of 1973-2010 at monthly scale. Since the goal model is multivariable, in addition to potential evapotranspiration data, relative humidity data, wind speed and sunshineare used to modeling the monthly evapotranspiration values. The models studied in this study are MPAR and MPAR-EGARCH models. The results of the verification and validation of the model data showed that both models are highly accurate. In this study, in all cases, the multivariate compilation model with conditional variance was more accurate than the multivariate periodic ARMA model. The results also showed that the MPAR-EGARCH compilation model fitted the minimum and maximum points of the studied data. The average error rate for estimating potential evapotranspiration values by MPAR model at stations of Birjand, Mashhad, Zabol and Zahedan was 0.4, 0.43, 1.05 and 3.04, respectively, and in the MPAR-EGARCH compilation models Respectively is equal to 0.16, 0.19, 0.55 and 0.59 respectively.