|
Abstract:
|
To perform hydrological forecasting, linear time series models are employed often. To explore the influence of the inflow on the outflow in a river system and exploit the internal interactions of the outflows, bivariate time series models are needed. The transfer function (TF) model and the semiparametric regression (SPR) model are used widely. In this paper, a new model---the nonparametric and functional-coefficient autoregressive (NFCAR)---is proposed. It consists of two parts: the nonparametric part, which explains the influences of the inflows on the outflow in a river system, and the functional-coefficient autoregressive, which reveals the interactions among the outflows in a river system. By comparing the calibration and forecasting of the models, it is found that the NFCAR model performs well.
|
