Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Pattimura International Journal of Mathematics (PIJMath)
Muhidin Jaariyah
Statistics Study Program, Faculty of Mathematics and Natural Sciences, Universitas Pattimura
Mathematics

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Forecasting The Composite Stock Price Index Using Autoregressive Integrated Moving Average Hybrid Model Artificial Neural Network Muhidin Jaariyah; Lexy Janzen Sinay; Norisca Lewaherilla; Yopi Andry Lesnussa
Pattimura International Journal of Mathematics (PIJMath) Vol 1 No 2 (2022): Pattimura International Journal of Mathematics (PIJMath)
Publisher : Pattimura University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (491.178 KB) | DOI: 10.30598/pijmathvol1iss2pp89-100

Abstract

A stock index is a statistical measure that reflects the overall price movement of a group of stocks selected based on certain criteria and methodologies and evaluated regularly. JCI is included in the composite index, which is the Headline index. The Headline Index is an index that is used as the main reference to describe the performance of the capital market. The JCI is very important in describing the current condition of the capital market because the JCI measures the price performance of all stocks listed on the Main Board and Development Board of the IDX. This study aims to predict JCI data using the time series method. The hybrid Autoregressive Integrated Moving Average–Artificial Neural Network (ARIMA-ANN) model combines the linear ARIMA model and the non-linear ANN model. The best models are the ARIMA model (2,1,1) and the ANN Backpropagation model with one input layer, one hidden layer with 20 neurons, and one output. The ARIMA-ANN hybrid model accurately predicts JCI data because it produces a MAPE value of less than 1%, with the level of forecasting accuracy from testing results being smaller than the level of accuracy during training. In addition, the forecast for the next five days is very accurate because it produces a very small RMSE and a MAPE below 1%, respectively, namely 56.99 and 0.72%.
Co-Authors Lexy Janzen Sinay Norisca Lewaherilla Yopi Andry Lesnussa