The Dynamic Regression model (DR) was used for forecasting the inflow to the Mosul and Dokan dams considering the effect of rainfall on the discharges.
The auto correlation function (ACF) was used to determine the stationary level of the time series. The partial auto correlation function (PACF) was used to identify a suitable ARIMA model for time series of rainfall and discharges for both dams. Besides, the other statistical tests: (χ2-test, t-test, RACF, Normality test, Trend analysis, CCF, Box-Cox method, Skewness Coefficient), and the factors of the transfer function models (TF) were determined. After passing the suitable statistical tests and the dynamic regression model for both dams, the model became suitable for forecasting the inflow depending upon the historical discharges and rainfall data.
Key words: Transfer Function – Dynamic Regression- Rainfall- Discharges- Modeling.