CD Skripsi
Prediksi pasar saham pada pt telkom indonesia (persero) tbk menggunakan algoritma long short terms memory
The stock market plays an important role as an indicator of business asset value and is the primary choice in the world of investment. PT Telkom Indonesia (Persero) Tbk is one of the state-owned enterprises whose shares are highly sought after by investors, but its share price often fluctuates due to economic, political, and market sentiment factors. The objective of this study is to predict the stock price of PT Telkom Indonesia (Persero) Tbk using the Long Short Term Memory (LSTM) algorithm. This study utilizes historical daily stock price data comprising 1,232 data points, spanning the period from January 1, 2019, to December 29, 2023. Data splitting in this study is 80% (Training) and 20% (Testing). The evaluation results using the Mean Absolute Percentage Error (MAPE) matrix yielded the following accuracy for each variable: Open (1.89%), High (1.25%), Low (1.69%), Close (1.25%), Adj Close (0.06%). This indicates that the LSTM model is highly effective, as a MAPE below 10% is categorized as very good accuracy. It can be concluded that the model successfully predicts the stock price of PT Telkom Indonesia (Persero) Tbk.
Keywords : Data Mining, LSTM, MAPE, Stock Market, PT Telkom Indonesia (Persero) Tbk
Tidak tersedia versi lain