臺灣博碩士論文加值系統

由於股票市場的複雜度，股價指數預測是一向非常艱巨的挑戰，需要投入較多的研究及資料去建構。文獻上大多以股票市場的資料來預測，而衍生性金融商品價格反映投資者對未來看法，且台灣的股票市場受機構法人投資行為影響，因此本文透過衍生性金融商品資料及長短期記憶神經網路來預測未來股價指數。

以期貨及選擇權資料，機構法人及大額交易者為組合，共建置21個變數，並以三個模型檢視機構法人投資行為是否能提升股價之預測能力，實驗結果發現機構法人及大額交易人期貨及選擇權的交易資料能提高未來股價指數之預測能力。

Due to stocks intricate behavior and unexpected movement, the stock price index forecasting has always entailed a very formidable challenge as it requires a compendium involving research and data analysis. Readily available literature dealt with stock market information, the price of derivative financial products reflect investors' views on the future, and Taiwan's stock market is affected by the investment behavior of institutional traders. Therefore, this study explores the use of derivatives market information such as futures and options with the utilization of Long-Short Term Memory neural network to predict the future stock price index.

By using futures and options data, institutional investors and large-value traders were combined to establish the 21 variables. Three predictive models were used to examine whether institutional investors’ views and behavior can improve and provide distinct contribution to the prediction ability of stock prices. Results of the study indicated that institutional investors and large-value traders, futures and options trading information can increase the accuracy of the predictive power of future stock price indices.

TABLE OF CONTENTS

ABSTRACT..............................................................i
ACKNOWLEDGMENTS ......................................................ii
TABLE OF CONTENTS.....................................................iv
LIST OF TABLES .......................................................v
LIST OF FIGURES .......................................................vi
CHAPTER 1: INTRODUCTION ...............................................1
1.1 Research Goals and Objectives..................................2
CHAPTER 2: LITERATURE REVIEW...........................................3
2.1 Artificial Intelligence: Stock Market Prediction...........3
2.2 Derivatives Market Information ............................6
CHAPTER 3: METHODOLOGY ...............................................9
3.1 Dataset ...............................................9
3.2 Definition of Variables....................................10
3.2.1 Futures..........................................11
3.2.2 Options..........................................12
3.2.3 Put-Call (PC) Ratio..............................12
3.2.4 Put-Call (PC) Power..............................13
3.2.5 Individual Ratio.................................13
3.3 Data Pre-Processing .......................................14
3.3.1 Normalization....................................14
3.3.2 The Architecture of the Data Process.............15
3.4 Neural Network: Long-Short Term Memory (LSTM)..............16
3.4.1 LSTM Parameters..................................17
3.5 Evaluation.................................................18
CHAPTER 4: EXPERIMENT AND EVALUATION...................................19
4.1 Experiment Setup...........................................19
4.2 Correlation Matrix.........................................19
4.3 Implementation Results.....................................21
CHAPTER 5: CONCLUSION..................................................26
REFERENCES.............................................................27

1. Abraham, A., Nath, B., & Mahanti, P.K. (2001). Hybrid intelligent systems for stock market analysis. In Computational Science-ICCS 2001 (pp.337-345). Springer.
2. Ahn, Jae Joon; Kim, Dong Ha; Oh, Kyong Joo; and Kim, Tae Yoon (2012). Applying option Greeks to directional forecasting of implies volatility in the options market: An intelligent approach. Expert Systems with Applications. Aug2012, Vol.39 Issue 10, p9315-9322. 8p.
3. Awrey, Dan (2016). The Mechanisms of Derivatives Market Efficiency. New York University Law Review. Nov 2016, Vol.91 Issue 5, p1104-1182. 79p.
4. Baek, Y. & Kim, H.Y. (2018). ModAugNet: A new forecasting framework for stock market index value with an overfitting prevention LSTM module and a prediction LSTM module. Expert Syst. Appl. 2018, 113, 457-480.
5. Chen, C.N., Chen, C.R., & Huang, Y.S. (2014). Which types of traders and orders profit from futures market trading? Journal of Derivatives, 21(4), 49-62.
6. Chen, H.C. & Wu, J. (2009). Volatility, depth, and order composition: Evidence from a pure limit order futures market. Emerging Markets Finance and Trade, 45(5), 72-85.
7. Chong, E., Han, C., & Park, F.C. (2017). Deep learning network for stock market analysis and prediction: Methodology, data representations, and case studies. Expert Syst. Appl. 2017, 83, 187-205.
8. Chou, H.C., Chen, W.N., & Chen, D.H. (2006). The expiration effects of stock-index derivatives – Empirical evidence from the Taiwan Futures Exchange. Emerging Markets Finance and Trade, 42(5), 81-102.
9. Chou, R.K. & Wang, Y.Y. (2009). Strategic order splitting, order choice, and aggressiveness: Evidence from the Taiwan Future Exchange. Journal of Futures Markets, 29(12), 1102-1129.
10. Chung, H. & Shin, K.-S. (2020). Genetic algorithm-optimized multi-channel convolutional neural network for stock market prediction. Neural Comput. Appl. 2020, 2, 7897-7914.
11. Erdinc Altay & Hakan Satman (2005). Journal of Financial Management and Analysis. 18(2), 18-33.
12. Garcia, F., Guijarro, F., Oliver, J. & Tamoŝiŭnienė, R. (2018). Hybrid fuzzy neural network to predict price prediction in the German DAX-30 index. Technol. Econ. Dev. Econ. 2018, 24, 2161-2178.
13. Goutam Datta, Pankaj Jha, Arnab Kumar Laha & Neeraj Mohan (2006). Journal of Emerging Market Finance, 5(3), 283-295.
14. Hsieh, W.L.G (2009). Expiration-day effects on individual stocks and the overall market: Evidence from Taiwan. Journal of Futures Markets, 29(10), 920-945.
15. Hsu, Tien-Yu (2021). Machine learning applied to stock index performance enhancement.Journal of Banking and Financial Technology. Published online on 25 January 2021. https://doi.org/10.1007/s42786-021-00025-6.
16. Huang, Y.C., & Chan, S.H. (2010). Trading behavior on expiration days and quarter-end days: The effect of a new closing method. Emerging Markets Finance and Trade, 46(4), 105-125.
17. Jeo Jasic & Douglas Wood (2004). Applied Financial Economics. 14, 285-297.
18. Jinyuan Shen, Huaiyu Fan & Shengjiang Chang (2007). Springer-Verlag Berlin Heidelberg, 457-464.
19. Kelotra, A., & Pandey, P. (2020). Stock Market Prediction Using Optimized Deep ConvLSTM Model. Big Data 2020, 8, 5-24.
20. Kyoung-jae, Kim & Won Boo Lee (2004). Neural Computation and Application, 13, 255-260.
21. Lee, Hsiu-Chuan; Chien, Cheng-Yi; Liao, Tzu-Hsiang (2012). Commonality in trading activity and future-cash basis: Evidence from the Taiwan futures and stock markets. Journal of Futures Markets. October 2012, Vol.32, Issue 10, pages 964-994.
22. Liu, F., & Wang, J. (2012). Fluctuation prediction of stock market index by Legendre neural network with random time strength function. Neurocomputing, 83, 12-21.
23. Mishra, R. K., Sehgal, S., & Bhanumurthy, N. (2011). A search for long-range dependence and chaotic structure in India stock market. Review of Financial Economics, 20, 96-104.
24. M. Nabipour, P. Nayyeri, H.Jabani,Shahab S. (2020). Predicting Stock Market Trends Using Machine Learning and Deep Learning Algorithms via Continuous and Binary Data; a Comparative Analysis. IEEE Access. Volume 8, 2020. Digital Object Identifier 10.1109/ACCESS.2020.3015966
25. Monica Isfan, Rui Menezes & Diana A, Mendes (2010). Journal on physics, 7th International Conference on Applications of Physics in Financial Analysis, 221, 1-14.
26. Pang, X., Zhou, Y., Wang, P., Lin, W., & Chang, V. (2018). An innovative neural network approach for stock market prediction. J. Supercomput.2018, 11, 1-21.
27. Qing Cao, Mark E. Parry & Karyl B. Leggio (2009). Springer Science and Business media, 250, 89-102.
28. Ramon Lawrence (1997). www.umanitoba.com, 1-21.
29. S.D. Bekiros & D.A Georgoutsos (2008). Journal of Forecasting. 19, 407-417.
30. Sharma, D.K, Hota,H.S., Handa, Richa (2021). Integration of genetic algorithm with artificial neural network for stock market forecasting. Int J Syst Assur Eng Manag. https://doi.org/10.1007/s13198-021-01209-5.
31. Taiwan Future Stock Exchange. Taiwan. Available at: http://www.taifex.com.tw
32. Taiwan Stock Exchange. Taiwan. Available at: http://www.twse.com.tw
33. TilaKaratne C.C., Mammadov M.A. & Morris S.A. (2009). Journal of Applied Mathematics and Decision Science. 1-22.
34. Vipul. (2005). Futures and options expiration-day effects: The Indian evidence. Journal of Futures Markets, 25(11), 1045-1065.
35. Vipul. (2008). Cross-market efficiency in the Indian derivatives market: A test of put-call parity. Journal of Futures Markets, 28(9), 889-910.
36. Xun Liang (2006). Springer-Verlag Berlin Heidelberg, 442-451.
37. Youngohc Yoon & George Swales (1991). Proceedings of the IEEE International Conference on Neural Networks, 156-162.
38. Zhou, X., Pan, Z., Hu, G., Tang, S., Zhao, C. (2018). Stock market prediction on high-frequency data using generative adversaries nets. Math. Probl. Eng. 2018, 2018.
39.https://www.investmentexecutive.com/news/research-and-markets/otc-derivatives-markets-growing-in-2019/