Please use this identifier to cite or link to this item: https://ah.nccu.edu.tw/handle/140.119/138002


Title: 以星狀生成對抗網路(STARGAN)解決股票量價聚合關係預測股票價值之研究:兼論以實驗計畫法調整超參數
Using Star Generative Adversarial Networks(StarGAN) to Resolve Joint Effect of Trading Volume and Price to Predict Stock Value – Finding Hyperparameters with Design of Experiment
Authors: 詹宗霖
JHAN, ZONG-LIN
Contributors: 姜國輝
Chiang, Kuo-Huie
詹宗霖
JHAN, ZONG-LIN
Keywords: 股價預測
量價關係
星狀生成對抗網路
深度學習
實驗計畫法
Stock Price Prediction
Star GAN
Deep Learning
Joint Effect
Design of Experiment Design
Date: 2021
Issue Date: 2021-12-01 14:29:48 (UTC+8)
Abstract: 股票市場中量與價具有聚合(joint)作用,但傳統之統計數值分析模型無法直接同時考慮兩者關係,僅能將其分開處理。本研究利用星狀生成對抗網路(Star GAN)多面向轉換的優點將證券的量價關係做結合,能夠建立模擬證券市場狀況的創新模型。我們繼續將Star GAN輸出的隱含量價之資料輸入常用於預測的LSTM模型預測未來1天或5天的交易量與價資料,達到股價預測之目的。在深度學中,超參數是一個難於解決之問題,我們採用田口實驗計畫法來確定最佳之超參數。實驗結果發現,使用25天的交易量與交易價資料當作輸入,預測1天後的股票資料效果最佳,在預測單一股票時誤差大約會在0.5~0.9%之間,而在預測多家公司股票時誤差會在0.4~0.6%之間。一般數值分析與機器學習方法,對於進行動態短預測有困難,本研究結合系統工程之動態系統來解決此一問題。實驗結果發現結合動態系統在單股預測之準確性(Accuracy)會有所提升,但在多股預測不會有明顯的提升,最重要的是會對預測精確性(Precision)明顯提升。整體而言,本研究運用Star GAN多面向轉換特性可以成功的處理證券量價關係以提升預測的準確性,並利用動態系統提升短期預測之準確率。
Joint effect exists between the volume and price of stock, but statistical stock analysis models cannot directly handle the joint effect but only consider the volume and price independently. This research uses the advantages of Star GAN, i.e. multi-faceted transformation, and uses result with the joint effect for the predictive model that can simulate the behavior of stock market. We feed the volume and price into Star GAN to obtain the output of potential volume and price, and then forward the results into LSTM model that is used for prediction of the volume and price in the next 1 or 5 days. However, there is great difficulty to predict the behavior of stock value in the short-term, thus. Further, Deep learning suffers from the difficulty of determination of the hyper-parameters. Thus, we apply Taguchi method for design of experimental to determine the optimal hyper-parameters. According to the experiment result, we found that using 25-day trading volume and trading price data as input to predict stock value for the next 1 day can obtain the best effect. The error lies between 0.5 and 0.9% by predicting a single stock and the error lies between 0.4 and 0.6% by predicting the prices of multiple company. It is also difficult to predict the stock price in short-term precisely, for that we introduce the dynamic system model in terms of system engineering to improve the precision of prediction. The experiment results reveal that it can improve precision of single-stock prediction, but cannot improve the precision of the multi-stock prediction. As a whole, this research use the multi-faceted transformation feature of Star GAN to improve the accuracy of stock price prediction and deal with the short-term stock price prediction for a single company with dynamic system successfully.
Reference: 1. Ying, C. C., (1966).Stock Market Prices and Volumes of Sales, Econometrica, 34, 676-685.
2. 林弈廷,(2019),以循環生成對抗網路預測股價量能動態關係,國立政治大學資管系碩士論文。
3. Lynn,從人工智慧、機器學習到深度學習,你不容錯過的人工智慧簡史,上網日期2020年3月10日,檢自: https://www.inside.com.tw/feature/ai/9854-ai-history
4. Andy Wang,Backpropagation(BP) 倒傳遞法 #1 工作原理與說明,上網日期2020年3月10日,檢自: https://www.brilliantcode.net/1326/backpropagation-1-gradient-descent-chain-rule/
5. Y. LeCun, B. Boser, J. S. Denker, D. Henderson, (1989). Backpropagation Applied to Handwritten Zip Code Recognition, Neural Computation, Volume 1, Issue 4, p.541-551.
6. G. E. Hinton*, R. R. Salakhutdinov,(2006). Reducing the Dimensionality of Data with Neural Networks, Science, Volume 313, Issue 5786, p.504-507.
7. Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, Yoshua Bengio,(2014). Generative Adversarial Networks, arXiv:1406.2661.
8. Qiao Zheng, Hervé Delingette, Nicolas Duchateau, Nicholas Ayache,(2018). 3D Consistent Biventricular Myocardial Segmentation Using Deep Learning for Mesh Generation, arXiv:1803.11080.
9. Jon Bruner, Generative Adversarial Networks for Beginners, 上網日期2020年3月13日,檢自:https://github.com/jonbruner/generative-adversarial-networks/blob/master/gan-notebook.ipynb
10. Jun-Yan Zhu*,Taesung Park*,Phillip Isola,Alexei A. Efros, Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks,上網日期2020年3月13日,檢自: https://junyanz.github.io/CycleGAN/
11. Jun-Yan Zhu, Taesung Park, Phillip Isola, Alexei A. (2017).Efros, Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks, arXiv:1703.10593.
12. Yunjey Choi, Minje Choi, Munyoung Kim, Jung-Woo Ha, Sunghun Kim, Jaegul Choo, (2018).StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 8789-8797.
13. DENNY BRITZ, Recurrent Neural Networks Tutorial, Part 1 – Introduction to RNNs, 上網日期2020年3月14日,檢自: http://www.wildml.com/2015/09/recurrent-neural-networks-tutorial-part-1-introduction-to-rnns/
14. Hochreiter, S., Schmidhuber, J., (1997)Long Short-Term Memory, Neural Computation, Volume 9, Issue 8, p.1735-1780.
15. Klaus Greff, Rupesh Kumar Srivastava, Jan Koutník, Bas R. Steunebrink, Jürgen Schmidhuber,(2017). LSTM: A Search Space Odyssey, IEEE Transactions on Neural Networks and Learning Systems ,Volume 28, Issue 10, p.2222-2232
16. Martin Sundermeyer, Ralf Schluter, and Hermann Ney, LSTM Neural Networks for Language Modeling, Human Language Technology and Pattern Recognition, Computer Science Department, RWTH Aachen University, Aachen, Germany.
17. 徐偉強,張永佳,李榮貴,(2011),基本面、籌碼面與總體經濟對台灣半導體產業股票報酬影響之研究,國立交通大學工業工程與管理學系碩士論文。
18. 蔡尚翰,(2017)籌碼面選股結合技術分析之投資績效研究,國立高雄應用科技大學資訊工程系碩士在職專班碩士論文。
19. 游英裕,股價與成交量因果關係之研究-台灣股市的實証,義守大學管理科學研究所碩士論文。
20. 陳仕偉,陳俊偉,(2006)台灣股票及外匯市場價量非線性因果關係之探討,Vol.2, No.1, p.21-51。
21. 杜芸菩,(2016)台灣八大類股價量關係,國立政治大學國際經營與貿易學系碩士論文。
22. 陳怡均,陳安斌,(2007)應用類神經網路對台股籌碼面與技術面之領先-落後研究分析,國立交通大學資訊管理研究所碩士論文。
23. 謝璁賦,陳安斌,(2010)應用類神經網路於台股權值股籌碼面的知識發現,國立交通大學資訊管理研究所碩士論文。
24. 郭裕凉,(2013)三大法人籌碼面預測臺灣加權股價指數之研究,國立高雄應用科技大學金融資訊研究所碩士論文。
25. 系統動力學-MBA智庫百科,上網日期2020年3月15日,檢自: https://wiki.mbalib.com/zh-tw/%E7%B3%BB%E7%BB%9F%E5%8A%A8%E5%8A%9B%E5%AD%A6
26. 田口品質工程-財團法人塑膠工業技術發展中心,上網日期2021年10月6日,檢自: https://www.pidc.org.tw/safety.php?id=124
27. 田口方法: 品質設計的原理與實務 第四版,李輝煌,上網日期2021年10月6號,檢自: https://web.archive.org/web/20200626172406/http://myweb.ncku.edu.tw/~hhlee/Myweb_at_NCKU/Taguchi4.html
28. 潘永浤,(2003) 應用田口方法於類神經網路輸入參數設計-零售商快速回應系統模式之建立為例,義守大學工業工程與管理學系碩士論文。
29. 田口品質工程,國立雲林科技大學工業工程與管理所品質與可靠度工程實驗室,上網日期2021年10月6號,檢自: https://www.iem.yuntech.edu.tw/lab/qre/public_html/source/DOE/files/%E7%94%B0%E5%8F%A3%E6%96%B9%E6%B3%95.pdf
30. 什麼是「布林通道」,CMoney 官方,上網日期2021年10月6號,檢自: https://www.cmoney.tw/learn/course/technicals/topic/1216
31. 黃郁佳,(2018) 布林通道交易策略之研究以台灣中型 100 指數為例,東海大學管理學院財務金融研究所碩士在職專班論文。
Description: 碩士
國立政治大學
資訊管理學系
107356017
Source URI: http://thesis.lib.nccu.edu.tw/record/#G0107356017
Data Type: thesis
Appears in Collections:[資訊管理學系] 學位論文

Files in This Item:

File Description SizeFormat
601701.pdf4060KbAdobe PDF0View/Open


All items in 學術集成 are protected by copyright, with all rights reserved.


社群 sharing