Generalized space time autoregressive with exogenous variable model and its application

Abstract

In this paper we proposed the Generalized Space Time Autoregressive with variable Exogenous, abbreviated GSTARX as GSTAR development with the addition of exogenous variables. GSTARX not only involves the element of time and location, but also the influence of exogenous variables in the model. GSTARX equation can be written as a linear model, so we can estimate parameters of GSTARX model using Ordinary Least Squares (OLS) method. For our case study, we use GSTARX model with uniform and inverse distance weights to predict an export volume of Crude Palm Oil (CPO) in several locations on the island of Sumatera, where X is the international CPO prices.