| dc.contributor.advisor | 曾正男 | zh_TW |
| dc.contributor.advisor | Tzeng, Jeng-Nan | en_US |
| dc.contributor.author (Authors) | 陳冠宇 | zh_TW |
| dc.contributor.author (Authors) | Chen, Guan-Yu | en_US |
| dc.creator (作者) | 陳冠宇 | zh_TW |
| dc.creator (作者) | Chen, Guan-Yu | en_US |
| dc.date (日期) | 2026 | en_US |
| dc.date.accessioned | 2-Mar-2026 13:38:57 (UTC+8) | - |
| dc.date.available | 2-Mar-2026 13:38:57 (UTC+8) | - |
| dc.date.issued (上傳時間) | 2-Mar-2026 13:38:57 (UTC+8) | - |
| dc.identifier (Other Identifiers) | G0110751008 | en_US |
| dc.identifier.uri (URI) | https://nccur.lib.nccu.edu.tw/handle/140.119/161933 | - |
| dc.description (描述) | 碩士 | zh_TW |
| dc.description (描述) | 國立政治大學 | zh_TW |
| dc.description (描述) | 應用數學系 | zh_TW |
| dc.description (描述) | 110751008 | zh_TW |
| dc.description.abstract (摘要) | 本研究探討幾何布朗運動模型（Geometric Brownian Motion, GBM）結
合模糊邏輯系統於股票交易策略預測之應用，並在一致的交易決策架構下，
比較蒙地卡羅模擬方法與深度學習模型（卷積神經網路CNN 與長短期記憶
網路LSTM）於不同預測情境中的交易績效表現。
研究中以GBM 為價格生成之理論基礎，透過回歸式參數更新機制（常
數、線性與二次回歸）建構滾動式預測模型，並引入模糊控制系統作為交易
決策層，將價格預測結果轉換為買入、賣出與持有等交易行為，以提升交
易策略之穩定性與可操作性。
本研究以台積電（TSMC）歷史股價資料進行跨年度模擬實驗，系統性
比較不同模型在固定預測期間、不同訓練視窗長度及資料更新設定下之表
現。實證結果顯示，在結合回歸式參數更新與模糊決策控制後，GBM 模型
於多數情境下之交易績效可與CNN 與LSTM 模型相當，顯示具理論基礎之
隨機模型在適當設計下仍具有實務應用價值。 | zh_TW |
| dc.description.abstract (摘要) | This study investigates the application of the Geometric Brownian Motion (GBM) model combined with a fuzzy logic system for stock trading strategy prediction, and conducts a systematic comparison with deep learning models, including Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM), under a unified trading decision framework.
GBM is employed as the theoretical foundation for price generation, with rolling parameter estimation based on constant, linear, and quadratic regression schemes. A fuzzy control system is introduced as the decision layer, transforming price predictions into trading actions such as buy, sell, and hold, thereby enhancing trading stability and practical applicability.
Using historical stock price data of Taiwan Semiconductor Manufacturing Company (TSMC), extensive simulation experiments are conducted across multiple years to evaluate model performance under different prediction horizons, training window lengths, and data updating settings. The empirical results indicate that, when combined with regression-based parameter updating and fuzzy decision control, the GBM-based models achieve trading performance comparable to that of CNN and LSTM models in many scenarios, demonstrating that theoretically grounded stochastic models remain competitive and practically valuable in modern trading applications. | en_US |
| dc.description.tableofcontents | 致謝...i
中文摘要...ii
Abstract...iii
目錄...iv
表目錄...viii
圖目錄...ix
第一章、引言...1
第二章、理論基礎與文獻探討...4
第三章、研究方法...7
第四章、實驗設計...26
第五章、模擬結果與分析...48
第六章、結論...85
參考文獻...90
附錄A、程式碼附錄...91 | zh_TW |
| dc.format.extent | 37268339 bytes | - |
| dc.format.mimetype | application/pdf | - |
| dc.source.uri (資料來源) | http://thesis.lib.nccu.edu.tw/record/#G0110751008 | en_US |
| dc.subject (關鍵詞) | 幾何布朗運動 | zh_TW |
| dc.subject (關鍵詞) | 蒙地卡羅模擬 | zh_TW |
| dc.subject (關鍵詞) | 模糊邏輯 | zh_TW |
| dc.subject (關鍵詞) | 股票預測 | zh_TW |
| dc.subject (關鍵詞) | 深度學習 | zh_TW |
| dc.subject (關鍵詞) | Geometric Brownian Motion | en_US |
| dc.subject (關鍵詞) | Monte Carlo Simulation | en_US |
| dc.subject (關鍵詞) | Fuzzy Logic | en_US |
| dc.subject (關鍵詞) | Stock Prediction | en_US |
| dc.subject (關鍵詞) | Deep Learning | en_US |
| dc.title (題名) | 應用幾何布朗運動模型與模糊系統於股票交易策略預測 | zh_TW |
| dc.title (題名) | Forecasting Stock Trading Strategies Using Geometric Brownian Motion Model and Fuzzy Systems | en_US |
| dc.type (資料類型) | thesis | en_US |
| dc.relation.reference (參考文獻) | [1] Shyi-Ming Chen. Forecasting enrollments based on fuzzy time series. Fuzzy Sets and Systems, 81(3):311–319, 2002.
[2] Paul Glasserman. Monte Carlo Methods in Financial Engineering. Springer, 2004.
[3] Ehsan Hadavandi, Hassan Shavandi, and Amir Ghanbari. Integration of genetic fuzzy systems and artificial neural networks for stock price forecasting. Knowledge-Based Systems, 23(8):800–808, 2010.
[4] James D. Hamilton. Time Series Analysis. Princeton University Press, 1994.
[5] John C. Hull. Options, Futures, and Other Derivatives. Pearson, 10 edition, 2018.
[6] Joel Lidén. Stock price predictions using a geometric brownian motion. Master’s thesis, Uppsala University, Uppsala, Sweden, 2018.
[7] Andrew W. Lo. Generalised geometric brownian motion: Theory and applications to option pricing. Journal of Financial Economics, 22(1):23–43, 1988.
[8] K. Reddy and V. Clinton. Comparison of stock price prediction using geometric brownian motion and multilayer perceptron. International Journal of Applied Engineering Research, 11(1):519–525, 2016.
[9] Ruey S. Tsay. Analysis of Financial Time Series. Wiley, 3 edition, 2010.
[10] Lotfi A. Zadeh. Fuzzy sets. Information and Control, 8(3):338–353, 1965. | zh_TW |
