Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/29008Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Muliono, Rizki | - |
| dc.contributor.author | xRiza, Evindo Amanda Riza | - |
| dc.date.accessioned | 2025-12-12T07:49:25Z | - |
| dc.date.available | 2025-12-12T07:49:25Z | - |
| dc.date.issued | 17-09 | - |
| dc.identifier.uri | https://repositori.uma.ac.id/handle/123456789/29008 | - |
| dc.description | 62 Halaman | en_US |
| dc.description.abstract | Saham merupakan salah satu instrumen investasi yang bersifat fluktuatif, sehingga diperlukan metode prediksi yang dapat mendukung pengambilan keputusan dalam aktivitas jual dan beli saham. Seiring dengan perkembangan teknologi kecerdasan buatan (AI), khususnya deep learning kini banyak dimanfaatkan untuk membantu investor dalam memahami pergerakan pasar. Penelitian ini bertujuan untuk membandingkan performa model deep learning, seperti Recurrent Neural Network (RNN) dan Long Short-Term Memory (LSTM) dalam memprediksi harga saham PT Bank Rakyat Indonesia Tbk (BBRI). Kedua model dilatih menggunakan kombinasi hyperparameter terbaik yang diperoleh melalui metode RandomizedSearchCV. Hasil penelitian ini menunjukkan bahwa model RNN menghasilkan nilai MSE sebesar 9.340,06, RMSE sebesar 96,64, dan MAE sebesar 76,57. Sementara itu, model LSTM menghasilkan nilai MSE sebesar 5.535,33, RMSE sebesar 74,39, dan MAE sebesar 57,25. Selain itu, hasil prediksi harga penutupan saham BBRI selama 7 hari ke depan menunjukkan bahwa RNN memiliki rata-rata persentase error sebesar 1,71%, sedangkan LSTM hanya sebesar 1,31%. Berdasarkan hasil tersebut, model LSTM terbukti memiliki performa yang lebih baik dibandingkan RNN dalam memprediksi harga penutupan saham BBRI, yang ditunjukkan oleh nilai error yang lebih rendah pada seluruh metrik evaluasi. Model prediksi diimplementasikan ke dalam sebuah aplikasi web menggunakan streamlit. Stocks are one of the investment instruments that are inherently volatile, thus requiring predictive methods to support decision-making in stock buying and selling activities. With the advancement of artificial intelligence (AI) technology, particularly deep learning, these methods are increasingly being utilized to help investors understand market movements. This study aims to compare the performance of deep learning models such as Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) in predicting the stock price of PT Bank Rakyat Indonesia Tbk (BBRI). Both models were trained using the best combination of hyperparameters obtained through the RandomizedSearchCV method. The results of this study show that the RNN model achieved a Mean MSE of 9.340,06, RMSE of 96,64, and MAE of 76,57. Meanwhile, the LSTM model achieved an MSE of 5.535,33, RMSE of 74,39, and MAE of 57,25. Additionally, the 7-day ahead stock closing price predictions indicate that the RNN model has an average percentage error of 1,71%, while the LSTM model has only 1,31%. Based on these results, the LSTM model proves to have better performance compared to the RNN model in predicting the closing price of BBRI stock, as indicated by the lower error values across all evaluation metrics. The predictive models were implemented into a web application using Streamlit. | en_US |
| dc.language.iso | id | en_US |
| dc.publisher | Universitas Medan Area | en_US |
| dc.relation.ispartofseries | NPM; | - |
| dc.subject | Saham | en_US |
| dc.subject | Prediksi Harga Saham | en_US |
| dc.subject | BBRI | en_US |
| dc.subject | RNN | en_US |
| dc.subject | LSTM | en_US |
| dc.subject | Streamlit | en_US |
| dc.subject | Stock | en_US |
| dc.subject | Stock Price Prediction | en_US |
| dc.title | Analisis Perbandingan Recurrent Neural Network dan Long Short Term Memory dalam Memprediksi Harga Saham | en_US |
| dc.title.alternative | Comparative Analysis of Recurrent Neural Network and Long Short Term Memory in Predicting Stock Prices | en_US |
| dc.type | Thesis | en_US |
| Appears in Collections: | SP - Informatic Engineering | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 218160018 - Evindo Amanda Riza - Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 2.25 MB | Adobe PDF | View/Open |
| 218160018 - Evindo Amanda Riza - Chapter IV.pdf Restricted Access | Chapter IV | 508.09 kB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.