Optimization of Autoregressive Integrated Moving Average (ARIMA) for Forecasting Indonesia Sharia Stock of Index (ISSI) using Kalman Filter

Luluk Wulandari, Yuniar Farida, Aris Fanani, Mat Syai'in

2018

Abstract

Forecasting stock price index is an important thing, because it can describe the state of stock price index going forward. It can be a consideration for a company's investors to determine the decision in selling, buying, or holding its stock. This research aims to find out an optimal model that can be used to forecast ISSI using ANN and ARIMA model. The optimal model was analyzed from the smallest RMSE and MAE results. The results of this research show that ANN (12,12,1) is more optimal than ARIMA (2, 1, 2) with values of MAE = 0.59143 and RMSE = 0.58705. Then, ARIMA model will be improved using Kalman filter method, showing that the residual value is very small with the RMSE value of 3.8693e-08. The RMSE value from the forecasting results using ARIMA – Kalman Filter is much smaller than the RMSE of ANN. Thus, it can be concluded that the ARIMA Kalman Filter method is more optimal than ANN in forecasting ISSI

Download


Paper Citation


in Harvard Style

Wulandari L., Farida Y., Fanani A. and Syai'in M. (2018). Optimization of Autoregressive Integrated Moving Average (ARIMA) for Forecasting Indonesia Sharia Stock of Index (ISSI) using Kalman Filter. In Proceedings of the Built Environment, Science and Technology International Conference - Volume 1: BEST ICON, ISBN 978-989-758-414-5, pages 295-303. DOI: 10.5220/0008906900002481


in Bibtex Style

@conference{best icon18,
author={Luluk Wulandari and Yuniar Farida and Aris Fanani and Mat Syai'in},
title={Optimization of Autoregressive Integrated Moving Average (ARIMA) for Forecasting Indonesia Sharia Stock of Index (ISSI) using Kalman Filter},
booktitle={Proceedings of the Built Environment, Science and Technology International Conference - Volume 1: BEST ICON,},
year={2018},
pages={295-303},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008906900002481},
isbn={978-989-758-414-5},
}


in EndNote Style

TY - CONF

JO - Proceedings of the Built Environment, Science and Technology International Conference - Volume 1: BEST ICON,
TI - Optimization of Autoregressive Integrated Moving Average (ARIMA) for Forecasting Indonesia Sharia Stock of Index (ISSI) using Kalman Filter
SN - 978-989-758-414-5
AU - Wulandari L.
AU - Farida Y.
AU - Fanani A.
AU - Syai'in M.
PY - 2018
SP - 295
EP - 303
DO - 10.5220/0008906900002481