Forecasting of Temperature and Rain in the City of Baghdad Using Seasonal Vector Autoregressive Moving Average Modelsvarma

Abstract

In this research , theinteraction  between two variable was studied and analyzed , the firstrepresenting the amount of monthly rainfall and the secondrepresenting the monthly temperature in Baghdad city , Abu Ghraib station for the period from Jan 2015 to Dec 2019 so it will be 60 values ,each variable represents a time series of 60 values , the two series were standardized with zero mean and one variance ,then the stationarity of the  two series was tested by the Diekey Fuller test, for the purpose of choosing the best order of the model , the  corrected Akaike information standard  AICC was calculated,It was found that the best model has the lowest value for the AICCstandard is SVARMA(2,0) , then the model was estimated , some tests were conducted including Portmanteau  test on the residual series of the estimated model,  AR disturbance test , and then forecast  for this model for a period of 24 months  , starting  from Jan 2020 to Dec 2021.The aim of this research is to analyze the relationship between rain and temperature with forecasting , due to importance of the amount of rainfall and the temperature on agriculture and the economy in general , it was concluded that The optimal model is SVARMA(2,0) ,the rain influenced by temperature at lag 1 , while temperature is influenced by temperature at lag 1 and 2 andthe variance increases as the forecast period increases . SAS programing was used to estimate and forecast the model .