Comparison of the TGARCH and EGARCH Model in Forecasting the Closing Stock Price of PT. Bank Central Asia Tbk

Closed

Anindya Hapsari Ariyoga, Lita Wulandari Aeli

2025 AIP Conference Proceedings Vol. 3446 Issue 1 Conference paper Cited by 0 Quartile

Abstract

Investment is the act of investing capital for future gains. One of the most popular investment instruments is stocks. In stocks there are various risks that occur, one of which is the stock price that goes up and down. The rise and fall of stock prices within a certain period of time will cause volatility, which is the existence of variance that is not constant or commonly referred to as heteroscedasticity. It is important for investors to know the situation of stock price developments to reduce the risks that occur such as panic buying and panic selling. One way to overcome this is by forecasting stock prices. The time series model commonly used for forecasting is Autoregressive Integrated Moving Average (ARIMA), but the ARIMA model is unable to overcome data containing heteroscedasticity effects. The model that can overcome the effects of heteroscedasticity is the Autoregressive Conditional Heteroscedasticity (ARCH)/Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model, but in stock price data tends to refer to bad news conditions (changes in stock prices down) which are greater than good news conditions (changes in stock prices up) which is commonly referred to as the asymmetric effect. The ARCH/GARCH model can only overcome symmetrical data, so this research is carried out forecasting using the GARCH model which can overcome asymmetrical effects. The forecasting models applied in this research are the TGARCH and EGARCH models, utilizing the closing stock prices of PT. Bank Central Asia Tbk as the dataset, covering the period from January 2, 2020, to March 4, 2024. The conclusion of this research is that the ARIMA(0,1,1)-TGARCH(1,1) model is better than ARIMA(0,1,1)-EGARCH(1,1). With SIC value of ARIMA (0,1,1) – TGARCH (1,1) is 12.19323 smaller than the SIC value of ARIMA(0,1,1) is 12.19994. The forecasted closing stock prices of PT. Bank Central Asia Tbk for the next four periods are Rp 9.767, Rp 9.799, Rp 9.932, and Rp 10.101. © 2025 American Institute of Physics Inc.. All rights reserved.

Affiliations

Department of Mathematics, Faculty of Mathematics and Natural Sciences, Universitas Negeri Malang, Malang, Indonesia