Application of Threshold Generalized Autoregressive Conditional Heteroscedastic (TGARCH) Model in Forecasting the LQ45 Stock Price Return

Jaka Nazarudin (Department of Mathematics, Faculty of Mathematics and Natural Science (FMIPA), Universitas Padjadjaran, Bandung, Indonesia)
Nurul Gusriani (Department of Mathematics, Faculty of Mathematics and Natural Science (FMIPA), Universitas Padjadjaran, Bandung, Indonesia)
Kankan Parmikanti (Department of Mathematics, Faculty of Mathematics and Natural Science (FMIPA), Universitas Padjadjaran, Bandung, Indonesia)
Sussy Susanti (Department of Management, Ekuitas School of Economic (STIE Ekuitas), Bandung, Indonesia)

Eksakta : Berkala Ilmiah Bidang MIPA
Vol. 24 No. 02 (2023): Eksakta : Berkala Ilmiah Bidang MIPA (E-ISSN : 2549-7464)

Economics is one of the most important fields for a country. One of the activities that illustrate the importance of the economy in a country is an investment. Investment activities, especially stock investment, are included in the capital market activities that various age groups currently carry out. Stocks are generally known to have high-risk, high-return characteristics. Therefore we need a way to minimize losses in investing. This study uses time series analysis theory to analyze LQ45 stock data.The data used is the closing price of PT. Bank Central Asia, Tbk. obtained from finance.Yahoo.com. The results of this study indicate that the return of daily closing price data of PT. Bank Central Asia, Tbk. during the period 2017-2021, there are heteroscedasticity and asymmetric shocks, so variations of the ARCH/GARCH model are needed to obtain accurate forecasting results. One suitable model is Threshold GARCH (TGARCH). The results of this study indicate that the suitable forecasting model for the data is the MA(3)-TGARCH(1,1) model. The model produces forecasts with an accuracy rate based on MAPE of 0.895% for the next seven days