Sanjaya Wisnu Ramadhan, - (2024) PREDIKSI HARGA SAHAM BERDASARKAN ANALISIS FUNDAMENTAL MENGGUNAKAN LONG SHORT-TERM MEMORY DAN LINEAR REGRESSION. S1 thesis, Universitas Pendidikan Indonesia.
Abstract
Investasi saham menjadi salah satu instrumen yang paling populer di kalangan masyarakat. Meskipun begitu, hadirnya berbagai faktor dan ketidakpastian membuat pasar saham sulit diprediksi. Salah satu cara dalam memprediksi harga saham adalah analisis fundamental. Analisis fundamental merupakan analisa yang dilakukan untuk mencari nilai wajar suatu saham. Data fundamental akan suatu perusahaan diambil berdasarkan kondisi saham seperti pada laporan finansial. Sehingga analisis fundamental berkaitan terhadap faktor rasional dan sebab-akibat terjadinya perubahan harga saham. Long Short-Term Memory (LSTM) dan Linear Regression merupakan model yang dapat membantu dalam melakukan prediksi harga saham. Penelitian ini mengevaluasi penerapan kedua model tersebut dalam memprediksi harga saham berdasarkan analisis fundamental. Faktor fundamental yang digunakan meliputi EPS, P/E, ROA, ROE, Debt/Equity, Market Cap, Price/Sales Ratio, Price/Book Ratio, Book Value Ratio, Total Assets Turnover Ratio, dan harga penutupan pada saham BBCA, TLKM, dan ASII. Hasil penelitian menunjukkan bahwa kedua model dapat diterapkan dengan baik dengan rasio pembagian dataset 80:10:10. Model LSTM menunjukkan nilai R2 sebesar 0,987 untuk BBCA, 0,814 untuk TLKM, dan 0,78 untuk ASII. Sementara itu, model Linear Regression menunjukkan hasil yang lebih baik dengan R2 masing-masing 0,994; 0,997; dan 0,994. Kesimpulan penelitian ini menunjukkan bahwa model Linear Regression lebih unggul dibandingkan LSTM dalam memprediksi harga saham berdasarkan analisis fundamental. Temuan ini memberikan kontribusi terhadap pengembangan strategi investasi yang lebih baik untuk mengurangi risiko dan meningkatkan keuntungan. ------------ Stock investment has become an important instrument among society. Even though, there is various factors and uncertainties makes the stock market difficult to predict. One way to predict stock prices is fundamental analysis. Fundamental analysis is an analysis carried out to find the fair value of a stock. The fundamental data about a company is taken based on stock condition such as financial reports. So fundamental analysis is related to rational factors and causes of changes in share prices. Long Short-Term Memory (LSTM) and Linear Regression are models that can help in predicting stock prices. The aim of this study is to evaluate both algorithm methods in predicting prices based on a fundamental analysis. Fundamental factors used include EPS (Earning per Share), P/E (Price to Earning), ROA (Return on Assets), ROE (Return on Equity), Debt/Equity, Market Cap, Price/Sales Ratio, Price/Book Ratio, Book Value Ratio, Total Assets Turnover Ratio, and closing prices on BBCA, TLKM and ASII share. Research results show that the two model can be well implemented with dataset split ratio of 80:10:10 Which gives the best results . LSTM model shows the R2 of 0,987 for BBCA, 0,814% for TKLM, and 0,78 for ASII. Then for the Linear Regression model shows a better R2 with 0,994, 0,997, and 0,994 for respective share. The conclusion of the results of this research shows that Linear Regression models are superior to LSTM in predicting share price based on fundamental analysis. these findings contribute to the development of better investment strategies to reduce risk and increase returns.
![[img]](http://repository.upi.edu/style/images/fileicons/text.png) |
Text
S_RPL_2003657_Title.pdf Download (736kB) |
|
Text
S_RPL_2003657_Chapter1.pdf Download (292kB) |
|
Text
S_RPL_2003657_Chapter2.pdf Restricted to Staf Perpustakaan Download (474kB) |
|
Text
S_RPL_2003657_Chapter2.pdf Restricted to Staf Perpustakaan Download (474kB) |
|
Text
S_RPL_2003657_Chapter3.pdf Download (698kB) |
|
Text
S_RPL_2003657_Chapter4.pdf Restricted to Staf Perpustakaan Download (2MB) |
|
Text
S_RPL_2003657_Chapter5.pdf Download (219kB) |
| Item Type: | Thesis (S1) |
|---|---|
| Additional Information: | https://scholar.google.com/citations?view_op=new_profile&hl=en ID SINTA Dosen Pembimbing: Raditya Muhammad: 6682222 Indira Syawanodya: 6681751 |
| Uncontrolled Keywords: | Investasi Saham, Analisis Fundamental, Long Short-Term Memory (LSTM), Linear Regression, Prediksi Harga, Stock Investment, Analysis Fundamental, Long Short-Term Memory (LSTM), Linear Regression, Price Prediction |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) |
| Divisions: | UPI Kampus cibiru > S1 Rekayasa Perangkaat Lunak |
| Depositing User: | Sanjaya Wisnu Ramadhan |
| Date Deposited: | 11 Sep 2024 03:07 |
| Last Modified: | 11 Sep 2024 03:07 |
| URI: | http://repository.upi.edu/id/eprint/121626 |
Actions (login required)
 |
View Item |