Computer Simulation Of Monthly Streamflow With Fgar(1) And Mgar(1) Models

Computer simulation is used to study many stochastic processes. Most of these stochastic processes in reality are generally skewed and dependent [5]. In the simulation of these processes, the first-order gamma autoregressive (GAR) (1) model has been found to be very effective; especially in the simulation of streamflow in Stochastic Hydrology. This paper mainly presents a study on the application of the Gar(1) model in the simulation of monthly streamflows. To achieve this aim we study the Gar(1) model and propose the FGar(1) and MGar(1) models. Based on the observed data of the monthly streamflows at the two stations and the series of monthly data for 1,000 years generated by means of computer simulation according to the FGar(1) and MGar(1) models, it was found that these models can reproduce the characteristic parameters of the historical data very well, particularly the mean and the standard deviation