Application of ARIMA Modeling to forecast the WeeklyStock Price per Share of (Persero) PT Telekomunikasi Indonesia Tbk

Mihardja, Jocelyn Beatrice

Journal of Actuarial, Finance, and Risk Managment
Vol 4, No 1 (2025)

This research is aimed to forecast the weekly stock price of PT Telkom using the ARIMA modeling approach. The dataset used in this study spanned from May 02, 2022 to January 30, 2023. The ARIMA(3,2,1) model was found to be the best fit for the data. The model's formula was Yt=-0.7692 Yt-1-7409 Yt-2-0.5175 Yt-3+et-0.2444 et-1, where Yt represents the stock price at time t and et represents the residual error at time t. This model was used to predict the stock price for the next 8 weeks. The accuracy of the chosen model is 85005.53, 0.0899, 353.67, and 8.99% respectively to MSE, RMSE, MAE, and MAPE. The forecasted results showed a gradual upward trend in the stock price with some fluctuations, indicating a positive outlook for PT Telkom in the coming weeks.