Investigation Of Seasonal Variation and Trend Effect in Stochastic Model Selection

Abstract

Stochastic models are frequently used methods in hydrology. These models are used to produce longer time period synthetic data having the same statistical parameters of limited observed data. Hydrologic data have usually seasonal variation and trend factors which effect to determine the best fit stochastic model. The effects of these parameters on various stochastic models were investigated in this study by using monthly averaged flow data of Demirci
(16-080) flow observation station in Konya basin in the present study. In various orders of AR, ARMA, SAR and SARMA models were evaluated for different seasonal effect and trend factor. In the second part of the study, several criteria were compared to determine the best fit model.