Investigating the performance of SARIMA linear time series and non-linear SETAR models in predicting monthly wind speed of
Authors
Abstract
In the present study, considering the importance of wind speed forecasting, one of the important factors especially in the discussion of evaporation in watershed, is the construction of water structures and dams ponds so For this purpose we use the SARIMA linear time series models and non-linear SETAR linear time series models to predict wind speed of Ahwaz synoptic station with a statistical period of 22 years 1988-2002. Correctly fitted Sarima models were selected based on self-correlation and partial correlation functions and independent residue test (Logon-Box). The SETAR models were also used to assess the adequacy of the remaining models of the model. The results of the evaluation of the matched models showed that the SETAR model (2; 6.2) with the coefficient of explanation of 0.74 compared to the SARIMA linear model had a precision It is acceptable to predict the monthly wind speed of Ahwaz synoptic station.
Keywords
