Predict of the monthly water volumes incoming in AL-ROOS River in the Syrian Coast by using the time series analysis


Abstract in English

The study and design of water dams depend essential on prediction of water volumes or future predicted in rivers, by using the time series analysis of the historical measurements. The research aims to make statistical study of monthly water volumes incoming in AL-Aroos River in Syrian coastal and future prediction of these volumes. And the Box-Jenkins models is adopt to analysis the time series data, because of its high accuracy. We attend the monthly water volumes for 15 years. And after doing the wanted tests on model residuals we found that the best model to represent the data is SARIMA(0,1,2) (1,2,1)12 , and after dividing the data to 14 years to build the model and one year to test it , and depending on the smallest of weighted mean of criteria RMSE, MAP, MAE,. The best predicted model is SARIMA (1,1,0) (0,1,1)12 and the model give the nearest predicted of measured data actually.

References used

THAFER, R. MUTTER, 2008, "Apposed technique for the problem of selecting the best forecasting model in time series: A case study", J. Iraqi Journal of Statistical Science., 1-20.IRAQ
SAWSAN, M. ALI, 2013, " Time series analysis of Baghdad rain fall using ARIMA method ", Iraqi journal of science, vol. 54 supplement No.4, pp:1136-1142, Baghdad –Iraq
SEVINC, O. Assessment ofclimate change effects in Aegean river basins: the case of Gediz and Buyuk Menderes basins.climatic change journal ISSN:0165 - 0009 (print) 1573-1480 (0nline), 2009, 253-283

Download