Using Semi-Parametric Approach to Estimate Amounts of Entering Water in Degla's River Model as a Class of Seasonal Integrated Fractional Processes

Abstract

Many studies have shown the existence of a pattern of time series that deals with the long-term processes, this process is referred to as long memory time series and that this type of model is verify the stationarity when taking the fractional differences (d) which lies within the closed period [-0.5, 0.5]. This study dealt with the mixed seasonal time series integrated fractionally through the semi-parametric estimation. Therefore, the aim of the research is to use of estimation methodology semi-parametric processes of the seasonal long-term using a number of segments of bandwidth on the seasonal frequencies through simulation Monte Carlo, with a comparison based on the mean squares error criterion. Finally, the amounts of water which flow Degla River in Iraq are used in our application.