Forecasting Rainfall at Surabaya using Vector Autoregressive (VAR) Kalman Filter Method

Yuniar Farida, Luluk Wulandari

2018

Abstract

Knowing the information of future rainfall data is necessary to increase awareness of the negative impacts of things caused by rainfall with high intensity to avoid loss and disaster. The aims of this research are forecasting rainfall at Surabaya city using Vector Autoregressive (VAR). This method is very simple because it is unnecessary to differentiate between variable of the dependent and independent. VAR is usually applied to the economic case and has optimal forecasting. But in this research will be applied to the weather case such as rainfall, humidity, temperature, and wind speed. The model used is the VAR (3) model. From the model, it is known that the value of R Square of rainfall is 0.56845. It shows that 56.845% model is influenced by the variable that defined in the model, the rest is influenced by other variables outside the model. Then obtained the forecast error of rainfall based on the MAPE value is 0.634581019. It shows that the residual value is high enough so that it needs to be improved using the Kalman Filter method. By applying Kalman Filter, it has decreased residual value very much. The MAPE value is become 0.008429293. So, the novelty of this research is VAR – Kalman Filter is very optimal to forecast weather such as rainfall, humidity, temperature, and wind speed which has fluctuative change.

Paper Citation

in Harvard Style

Farida Y. and Wulandari L. (2018). Forecasting Rainfall at Surabaya using Vector Autoregressive (VAR) Kalman Filter Method.In Proceedings of the International Conference on Mathematics and Islam - Volume 1: ICMIs, ISBN 978-989-758-407-7, pages 342-349. DOI: 10.5220/0008521703420349

in Bibtex Style

@conference{icmis18,
author={Yuniar Farida and Luluk Wulandari},
title={Forecasting Rainfall at Surabaya using Vector Autoregressive (VAR) Kalman Filter Method},
booktitle={Proceedings of the International Conference on Mathematics and Islam - Volume 1: ICMIs,},
year={2018},
pages={342-349},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008521703420349},
isbn={978-989-758-407-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the International Conference on Mathematics and Islam - Volume 1: ICMIs,
TI - Forecasting Rainfall at Surabaya using Vector Autoregressive (VAR) Kalman Filter Method
SN - 978-989-758-407-7
AU - Farida Y.
AU - Wulandari L.
PY - 2018
SP - 342
EP - 349
DO - 10.5220/0008521703420349