Pemodelan Data Curah Hujan dengan Model Generalized Space Time Autoregressive (GSTAR)

Abstract

Space Time Autoregressive (STAR) is a model that can be used to analyze time series data that havea relationship between time and location. However STAR models tend to be inflexible when faced with locationthat have heterogeneous characteristic. To overcome this weaknes, the Generalized Space Time Autoregressive(GSTAR) model is used, which is a STAR model, assuming that the autoregressive parameters and space-timeparameters don’t have be the same value at each location. The purpose of this study is to modeling rainfalldata in Pringsewu Regency(Z1), South Lampung Regency(Z2), and South Lampung Regency using the GSTARmodel. By using Distance Invers weighting matrix, the model obtained is GSTAR (31) I(0).