ARMA Modeling of a Stochastic Process Appropriate for the Angat Reservoir

Abstract

Modeling a hydrologic time series is generating a synthetic sequence that can be used in the operation of a water resource system. Most time series are stochastic in that future values are only partly determined by past values. The paper presented the ARMA model selection process that was used by this author in a previous study and showed that the selected autoregressive model of the Angat Reservoir inflows was appropriate by comparing the generated model outputs with recent observed measurements from 1986 to 2008.

Keywords: autoregressive model, moving-average model, Angat Reservoir inflows, streamflow forecasting, ARMA modeling, hydrologic model