Comparative Analysis of EGARCH and TGARCH Models in Stock Price Prediction

Arya Impun Diapari Lubis (Fakultas Sains dan Teknologi UIN Sumatera Utara)
Sajaratud Dur (Fakultas Sains dan Teknologi UIN Sumatera Utara)
Ismail Husein (Fakultas Sains dan Teknologi UIN Sumatera Utara)

Zero : Jurnal Sains, Matematika, dan Terapan
Vol 6, No 1 (2022): Zero: Jurnal Sains Matematika dan Terapan

Stocks are proof of the value of ownership of a company which are usually sold on the capital market, companies that buy and sell their shares will be easy to find with the existence of the stock market. The fund obtained by the company from investors who invest in several companies. Investors need to understand the models valuation of stock prices because investors have interest with changes in share prices. The purpose of study for looking the difference of the EGARCH model with TGARCH as a comparison which one is better at predicting stock prices. This research is a quantitative study using the EGARCH and TGARCH models by use Quasi Maximum Likelihood (QML) method. It was found that ARIMA (1 0 1) EGARCH (3 4) is a model that shows the best performance based on the smallest AIC value and the significance of all parameters. The ARIMA (1 0 1) EGARCH (3 4) model formed for forecasting returns and volatility is as follows: with ARIMA (1 0 1) EGARCH (3 4) models also have the MAE (Mean Absolute Error) value is 0.044%.