Modeling of Monthly Mean Values of Temperature on the Basis of SARIMA

Temperature is an important parameter, which contributes cliinate conditions effect of temperature increasing and decreasing on transportation, farming upwelling, down welling, hunting, building, and suchlike is very iinportant. In this assay with 53 years of monthly mean values of Bushehr synoptic station temperature (1331-1383) on the following basis was simnulated. 1-Autocorrelation function (ACF) and partial autocorrelation function (PACF), 2- Examine whether or not the residuals follow a normal distribution by an Anderson-Darling test, 3-Comparing the models with Akaike information criterion (AIC) and Schwarz Bayesian criterion (SBC). Time series model, the monthly mean SARIMA (1,0,2)(0,1, with the applied values of temperature predicted. On the basis of suggested model we can forecast mean values of temperature with the high accuracy and confidence limits.

Keywords:

Autocorrelation , Partial Autocorrelation , Model , Time Series , Stationary , Residuals