Penerapan Model Indeks Tunggal dalam Menganalisis Portofolio Saham Optimal dan Peramalan Harga Saham Optimal Menggunakan Long Short-Term Memory

Abstract

Investors in choosing stocks are not only on one type of stock, but can diversify. In this study, Blue Chip stocks were used which was carried out in two stages. The first stage is the Single Index Model to obtain an optimal stock portfolio. The second stage, the optimal share price results from the Single Index Model, namely Bank Mandiri (Persero) Tbk (BMRI), is followed by a forecasting process using Long Short-Term Memory (LSTM). The purpose of this study is to apply the Single Index Model to analyze the portfolio on Blue Chip stocks, forecast the selected optimal stock price in the next day, and evaluate the results of the selected optimal stock price forecasting. In the first stage, the Single Index Model uses data for the period from February 1, 2021 to February 1, 2024. The second stage, forecasting with LSTM uses data for the period from February 1, 2021 to February 29, 2024. The results of the first stage were obtained with optimal shares of 13 shares out of 41 Blue Chip shares. The most optimal stock was obtained by BMRI shares. The second stage is the use of BMRI shares with LSTM forecasting. The results of the second stage were obtained by forecasting the stock price performance with the best performance, so that it shows that the stock price performance on March 1, 2024 was obtained of IDR 6894,574. The accuracy of forecasting stock price performance with MAPE is 4.23%. It can be said that the forecast of stock price performance has very good accuracy.