Egarch Model Prediction for Sale Stock Price

Abstract

Stock is an investment in the capital market that is very promising for investors. Investors can also get high returns from the shares invested. However, this stock price is not always stable, it can go up and down drastically. The purpose of this study is to predict stock prices because they often experience instability. The method used in this research is using the Exponential generalized autoregressive conditional heteroscedasticity (EGARCH) model with the Quasi Maximum Likelihood (QML) method. The result of this research is the implementation of this model. The EGARCH model used is the stock price index model that is formed, namely the autoregressive integrated moving average (ARIMA) (0, 1, 2) EGARCH (1.4). The conclusion from the results of the research that predictions using the ARIMA model (0, 1, 2) EGARCH (1, 4) is the best model in accommodating the asymmetric nature of the volatility of the stock price index. The results of this egarch model show more optimal prediction results seen from an error of 3% compared to other modes such as the arch model and the GARCH model.

Peramalan Volatilitas IHSG dan Estimasi Value-at-Risk Menggunakan Model Student APARCH

“…Beberapa penelitian fokus pada peramalan volatilitas indeks luar negeri seperti S&P500 (Hafizah dkk., 2020;Megawati dkk., 2020). Sebagian penelitian lain berusaha untuk menelaah volatilitas saham individual (Husein & Lubis, 2022;Maulana, 2020;Sidadolog dkk., 2020), ataupun volatilitas pada sektor tertentu yang merupakan subkomponen dari IHSG (Ridha & Wibowo, 2020). Selain pada pasar saham, penelitian mengenai peramalan volatilitas juga telah dilakukan pada pasar kripto (Warsito & Robiyanto, 2020).…”

“…Selain pada pasar saham, penelitian mengenai peramalan volatilitas juga telah dilakukan pada pasar kripto (Warsito & Robiyanto, 2020). Ditinjau dari model yang digunakan, penelitian-penelitian tersebut menggunakan Asymmetric Power ARCH (APARCH) (Sidadolog dkk., 2020), IGARCH (Megawati dkk., 2020), Threshold GARCH (TGARCH) (Ridha & Wibowo, 2020), EGARCH (Husein & Lubis, 2022;Maulana, 2020), dan GJR-GARCH (Maulana, 2020). Hanya saja penelitian-penelitian tersebut tidak memprediksi volatilitas sebagai upaya untuk menghitung value-at-risk yang merupakan salah satu penerapan dari model prediksi volatilitas.…”