Modeling and Forecasting of Rainfall in Bengkulu City using Generalized Space-Time Autoregressive (GSTAR)
Herlin Fransiska (a*), Etis Sunandi (a), Dian Agustina (a)
a) Dept. of Statistics, University of Bengkulu.
*hfransiska[at]unib.ac.id
Abstract
Rainfall is an essential problem in Bengkulu City. Heavy rainfall can have bad impacts such as floods and landslides. This study aims to model monthly rainfall data in Bengkulu City by considering location factors. These locations are the coast, lowlands, and highlands so that they have different rainfall. Modeling using space-time data using the Generalized Space-Time Autoregressive (GSTAR) method. Rainfall data used monthly rainfall data from 2008 to 2017. The data source is BMKG. Stages of analysis are model identification, estimation of model parameters, and model validation. The best model can be used as a reference and input for climate and weather research in Bengkulu for the formulation of policies in disaster management by local governments through the National Disaster Management Agency (BNPB) for Bengkulu City.
Keywords: Bengkulu City, GSTAR, Rainfall, Space-Time Data
Topic: Mathematics