Introduction
In long or short-term river operation, river flow estimation is an important parameter. One of the common methods employed is based on using past observed data and forecasting river discharge in the future or using time series analysis. The field of time series analysis has changed in the last decade due to progress and acquisition of new knowledge in non-linear dynamics (Sprave, 1994). Nevertheless, there are still applications where the accurate estimation of linear
processes such as auto regressive moving average (ARMA) models are sufficient especially when they are used for linear time series analysis (Hwarng, 2001). However, the methods for this class of models were developed more than 20 years ago, with the restrictions of the then currently available computer resources. Therefore, it is necessary to test the new approaches in applying the ARMA models in time series analysis.