Statistical methods for river runoff prediction

Abstract

Methods used to analyze one type of nonstationary stochastic processes-the periodically correlated process-are considered. Two methods of one-step-forward prediction of periodically correlated time series are examined. One-step-forward predictions made in accordance with an autoregression model and a model of an artificial neural network with one latent neuron layer and with an adaptation mechanism of network parameters in a moving time window were compared in terms of efficiency. The comparison showed that, in the case of prediction for one time step for time series of mean monthly water discharge, the simpler autoregression model is more efficient.

Towards landslide predictions: two case studies

“…Frequency doubling or more generally frequency multiplications are the results of simple nonlinearities. Indeed, higher frequency overtones in river runoff is very common feature of hydrological regime [28]. In contrast, the creation of sub-harmonics requires bifurcations or period-doubling, for instance involving nonlinear processes with time delays.…”

Forecasting by Stochastic Models to Inflow of Karkheh Dam at Iran

“…Also the river runoff forecast based on the modeling of time series used to by Nigam et al (2014) [10]. Pisarenko et al (2005) used the Statistical methods for river runoff prediction and the result showed that, in the case of prediction for one time step for time series of mean monthly water discharge, the simpler auto-regression model is more efficient [12].…”