Febriyanti, Alvi Yuana (2025) PREDIKSI HARGA SAHAM DAN RISIKO KERUGIAN MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN) DAN LONG SHORT-TERM MEMORY (LSTM). Undergraduate thesis, UNIVERSITAS PEMBANGUNAN NASIONAL "VETERAN" JAWA TIMUR.
![]() |
Text (Cover)
21083010030_Cover.pdf Download (1MB) |
![]() |
Text (Bab 1)
21083010030_Bab 1.pdf Download (242kB) |
![]() |
Text (Bab 2)
21083010030_Bab 2.pdf Restricted to Registered users only until 19 June 2028. Download (615kB) |
![]() |
Text (Bab 3)
21083010030_Bab 3.pdf Restricted to Registered users only until 19 June 2028. Download (391kB) |
![]() |
Text (Bab 4)
21083010030_Bab 4.pdf Restricted to Registered users only until 19 June 2028. Download (6MB) |
![]() |
Text (Bab 5)
21083010030_Bab 5.pdf Download (211kB) |
![]() |
Text (Daftar Pustaka)
21083010030_Daftar Pustaka.pdf Download (191kB) |
![]() |
Text (Lampiran)
21083010030_Lampiran.pdf Restricted to Registered users only until 19 June 2028. Download (4MB) |
Abstract
Stock price prediction is a major challenge in the financial domain due to high volatility and complex movement patterns. Traditional methods such as fundamental and technical analysis often fail to capture the non-linear characteristics and rapidly changing dynamics of the market, highlighting the need for a more adaptive approach. This study proposes a hybrid deep learning model, CNN-LSTM, which combines the local feature extraction capabilities of CNN with the long-term temporal dependency modeling capabilities of LSTM. To incorporate risk management, this model is also integrated with the Value at Risk (VaR) approach using the Cornish-Fisher Expansion (ECF) to estimate potential losses under extreme market conditions as a novelty/innovation in this research. This research update also adds a GUI to implement the model. This research uses historical daily stock price data from PT Unilever Indonesia Tbk obtained from Yahoo Finance. Model performance is evaluated using Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE), where the model achieves an MAE of 0.049978 and a MAPE of 5.24%, indicating relatively low absolute and relative prediction errors. These results confirm that the CNN-LSTM approach effectively models stock price movements in a dynamic market environment, and integration with VaR-ECF provides risk estimates at a 95% confidence level of -0.0111 and 99% confidence level of -0.0166. Thus, this approach not only improves prediction accuracy but also offers valuable decision-support tools for investors in planning investment strategies. Keywords: Stock Price Prediction, CNN-LSTM, Value at Risk (VaR), Cornish Fisher Expansion (ECF), Flask.
Item Type: | Thesis (Undergraduate) | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Contributors: |
|
||||||||||||
Subjects: | H Social Sciences > HA Statistics Q Science > Q Science (General) Q Science > QA Mathematics T Technology > T Technology (General) Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science |
||||||||||||
Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
Depositing User: | Alvi Yuana Febriyanti | ||||||||||||
Date Deposited: | 20 Jun 2025 01:34 | ||||||||||||
Last Modified: | 20 Jun 2025 01:34 | ||||||||||||
URI: | https://repository.upnjatim.ac.id/id/eprint/38554 |
Actions (login required)
![]() |
View Item |