Selayanti, Nabilah (2026) Implementasi Model TCN-GAN untuk Prediksi Harga Saham dan Risiko Kerugian Pada Investasi Saham Emiten Crude Palm Oil (CPO). Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
22083010013-cover-3-20.pdf Download (1MB) |
|
Text (Bab 1)
22083010013-bab1.pdf Download (189kB) |
|
Text (Bab 2)
22083010013-bab2.pdf Restricted to Repository staff only until 9 March 2029. Download (1MB) | Request a copy |
|
Text (Bab 3)
22083010013-bab3.pdf Restricted to Repository staff only until 9 March 2029. Download (795kB) | Request a copy |
|
Text (Bab 4)
22083010013-bab4.pdf Restricted to Repository staff only until 9 March 2029. Download (2MB) | Request a copy |
|
Text (Bab 5)
22083010013-bab5.pdf Download (136kB) |
|
Text (Daftar Pustaka)
22083010013-daftarpustaka.pdf Download (161kB) |
|
Text (Lampiran)
22083010013-lampiran.pdf Restricted to Repository staff only until 9 March 2029. Download (116kB) | Request a copy |
Abstract
Crude Palm Oil (CPO) industry is a strategic sector for Indonesia's economy due to its contribution to national exports and the capital market. However, CPO prices, influenced by global market dynamics and export policies, cause the stock prices of companies in this sector to be highly volatile and risky, posing challenges in producing accurate stock price predictions and comprehensive investment risk assessments. The urgency for price prediction and risk analysis increases as the number of active investors in the Indonesian capital market grows. Therefore, an approach was carried out to improve the accuracy of stock price predictions while estimating potential losses to support investment decision-making. The innovation of this study lies in the integration of stock price predictions from the TCN-GAN model, with ARIMA as a comparison, into investment loss risk analysis. The study aims to implement the best prediction model to estimate investment loss risk using the Value at Risk (VaR) method with Historical Simulation (HS) based on predicted stock prices. The data used consist of daily closing prices of PT PP London Sumatra Indonesia Tbk. (LSIP.JK) and PT Sawit Sumbermas Sarana Tbk. (SSMS.JK) obtained from Yahoo Finance. The results show that TCN-GAN outperforms ARIMA in prediction accuracy. For LSIP.JK, TCN-GAN achieves a MAPE of 2.34% and RMSE of 35,08, while ARIMA produces a MAPE of 23,23% and RMSE of 319,46. For SSMS.JK, TCN-GAN achieves a MAPE of 5,02% and RMSE of 192,05, whereas ARIMA has a MAPE of 18,11% and RMSE of 368,68. Therefore, TCN-GAN was used to predict the next five periods, with average predicted prices of Rp1.363,13 and Rp1.574,97, respectively. Value at Risk (VaR) analysis at a 95% confidence level and a five-day holding period indicates a maximum potential loss of Rp56.859 dan Rp59.441.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | H Social Sciences > HA Statistics Q Science > Q Science (General) Q Science > QA Mathematics > QA76.6 Computer Programming T Technology > T Technology (General) |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Nabilah Selayanti | ||||||||||||
| Date Deposited: | 09 Mar 2026 07:44 | ||||||||||||
| Last Modified: | 09 Mar 2026 08:15 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/50217 |
Actions (login required)
 |
View Item |