近年來人工智慧 (Artificial Intelligence, AI) 在金融領域的應用愈來愈廣泛，其中應用在資產定價也是個熱門話題。本研究將以台灣加權指數為標的之選擇權日內資料用機器學習(Machine Learning, ML)做選擇權定價，以此方式和傳統的Black – Scholes選擇權定價模型做比較，觀察結果是否更加貼近實際的市場價格。本研究也在實證中做了訓練目標為選擇權價格與波動度的模型比較，和不同資料頻率的模型比較。實證結果顯示，用機器學習做選擇權定價能比BS模型更貼近實際價格，而訓練目標為波動度的表現優於選擇權價格，頻率越高的資料定價能力也越高。不過就預測未來半年價格的能力而言，機器學習的預測能力沒有比BS模型直接用前一期價格做還好。

In recent years, the application of artificial intelligence in the field of finance has become more and more widespread. Among them, the application in asset pricing is also a hot topic. In this study, I use machine learning (ML) for option pricing with intraday data of Taiwan Capitalization Weighted Stock Index (TAIEX). This method is compared with the traditional Black-Scholes option pricing model to see if the results are closer to the actual market price. This study also empirically compares models with output of option price and volatility, and models with different data frequencies.The empirical results show that using machine learning for option pricing can be closer to the actual price than the BS model, and the training objective of volatility outperforms the option price, and the higher the frequency, the higher the information pricing ability. However, in terms of the ability to predict prices for the next six months, machine learning is no better than a BS model that uses the previous period's prices directly.

Contents
論文審定書i
摘要ii
Abstractiii
目錄iv
List of Figuresv
List of Tablesvi
1. Introduction 1
1.1 Background 1
1.2 Research Purpose 2
1.3 Thesis Organization 3
2. Literature Review 5
2.1 Option Pricing Model 5
2.2 XGBoost and CatBoost for Financial Applications 9
2.3 Option Pricing Using Machine Learning 11
3. Methodology 15
3.1 Research Design 15
3.2 Data 16
3.3 Model Description 17
3.4 Model Comparisons 25
4. Empirical Results 30
4.1 Descriptive Statistics of Data 30
4.2 Model Comparisons 31
5. Conclusions 38
References 40
List of Figures
FIGURE 1 RESEARCH FRAMEWORK 4
FIGURE 2 FLOW CHART 15
FIGURE 3 FIRST STAGE OF MODEL COMPARISON 26
FIGURE 4 SECOND STAGE OF MODEL COMPARISON 28
FIGURE 5 ROLLING SCHEME DIAGRAM 28
FIGURE 6 RECURSIVE SCHEME DIAGRAM 28
FIGURE 7 RECURSIVE SCHEME DIAGRAM 29
 
List of Table
TABLE 1 FEATURES SETS TABLE 27
TABLE 2 DESCRIPTIVE STATISTICS OF THE IMPORTANT VARIABLES 31
TABLE 3 COMPARISON OF DIFFERENT LABEL AND FEATURES 32
TABLE 4 FIRST SET OF FEATURES CUT BY ROLLING SCHEME 33
TABLE 5 FIRST SET OF FEATURES CUT BY RECURSIVE SCHEME 34
TABLE 6 FOURTH SET OF FEATURES CUT BY ROLLING SCHEME 34
TABLE 7 FOURTH SET OF FEATURES CUT BY RECURSIVE SCHEME 35
TABLE 8 COMPARISON OF DIFFERENT DATA FREQUENCIES
36

Abdelmalek, W., Hamida, S. B., & Abid, F. (2009). Selecting the best forecasting-implied volatility model using genetic programming. Advances in Decision Sciences, 2009.
Al Daoud, E. (2019). Comparison between XGBoost, LightGBM and CatBoost using a home credit dataset. International Journal of Computer and Information Engineering, 13(1), 6-10.
Andreou, P. C., Charalambous, C., & Martzoukos, S. H. (2009, September). European option pricing by using the support vector regression approach. In International Conference on Artificial Neural Networks (pp. 874-883). Springer, Berlin, Heidelberg.
Basak, S., Kar, S., Saha, S., Khaidem, L., & Dey, S. R. (2019). Predicting the direction of stock market prices using tree-based classifiers. The North American Journal of Economics and Finance, 47, 552-567.
Berkowitz, J. (2009). On justifications for the ad hoc Black-Scholes method of option pricing. Studies in Nonlinear Dynamics & Econometrics, 14(1).
Black, F., & Scholes, M. (1973). The pricing of options and corporate liabilities. Journal of political economy, 81(3), 637–654.
Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Paper presented at the Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining.
Choi, Y., Jordan, S. J., & Ok, S. (2012). Dividend‐rollover effect and the ad hoc Black‐Scholes model. Journal of Futures Markets, 32(8), 742-772.
Chowdhury, R., Mahdy, M. R. C., Alam, T. N., Al Quaderi, G. D., & Rahman, M. A. (2020). Predicting the stock price of frontier markets using machine learning and modified Black–Scholes Option pricing model. Physica A: Statistical Mechanics and its Applications, 555, 124444.
Culkin, R., & Das, S. R. (2017). Machine learning in finance: the case of deep learning for option pricing. Journal of Investment Management, 15(4), 92-100.
Dumas, B. J., Fleming, J., & Whaley, R. E. (1998). Implied volitility functions: Empirical tests. Journal of finance, 53, 2059-2106.
Han, H., Huang, H., Hu, J., & Kuang, F. (2019, June). Implied Volatility Pricing with Selective Learning. In International conference on Data Science, Medicine and Bioinformatics (pxp. 18-34). Springer, Singapore.
Heston, S. L. (1993). A closed-form solution for options with stochastic volatility with applications to bond and currency options. The review of financial studies, 6(2), 327-343.
Hull, J., & White, A. (1987). The pricing of options on assets with stochastic volatilities. The journal of finance, 42(2), 281-300.
Itkin, A. (2019). Deep learning calibration of option pricing models: some pitfalls and solutions. arXiv preprint arXiv:1906.03507.
Ivașcu, C. F. (2021). Option pricing using machine learning. Expert Systems with Applications, 163, 113799.
Jabeur, S. B., Gharib, C., Mefteh-Wali, S., & Arfi, W. B. (2021). CatBoost model and artificial intelligence techniques for corporate failure prediction. Technological Forecasting and Social Change, 166, 120658.
Jeremy Berkowitz. (2009).On Justifications for the ad hoc Black-Scholes Method of
Option Pricing. Department of Finance University of Houston.
Jha, S. B., Babiceanu, R. F., Pandey, V., & Jha, R. K. (2020). Housing Market Prediction Problem using Different Machine Learning Algorithms: A Case Study. arXiv preprint arXiv:2006.10092.
Kim, S., & Lee, C. (2014). On the importance of the traders’ rules for pricing options: Evidence from intraday data. Asia‐Pacific Journal of Financial Studies, 43(6), 873-894.
Kim, S. (2021). Portfolio of Volatility Smiles versus Volatility Surface: Implications for pricing and hedging options. Journal of Futures Markets, 41(7), 1154-1176.
Liang, X., Zhang, H., Xiao, J., & Chen, Y. (2009). Improving option price forecasts with neural networks and support vector regressions. Neurocomputing, 72(13-15), 3055-3065.
Liudmila Prokhorenkova, Gleb Gusev, Aleksandr Vorobev, Anna Veronika Dorogush, Andrey Gulin (2018). “CatBoost: unbiased boosting with categorical features”. Advances in Neural Information Processing Systems 31 (NeurIPS 2018).
Liu, S., Oosterlee, C. W., & Bohte, S. M. (2019). Pricing options and computing implied volatilities using neural networks. Risks, 7(1), 16.
Merton, R. C. (1973). Theory of rational option pricing. The Bell Journal of economics and management science, 141-183.
Ndikum, P. (2020). Machine learning algorithms for financial asset price forecasting. arXiv preprint arXiv:2004.01504.
Wang, C. W., Wu, C. W., Tzang, S. W. (2012). Implementing option pricing models when asset returns follow an autoregressive moving average process. International Review of Economics and Finance 24 (2012) 8-25.
Yung, H. H., & Zhang, H. (2003). An empirical investigation of the GARCH option pricing model: Hedging performance. Journal of Futures Markets: Futures, Options, and Other Derivative Products, 23(12), 1191-1207.
Zeng, Y., & Klabjan, D. (2019). Online adaptive machine learning based algorithm for implied volatility surface modeling. Knowledge-Based Systems, 163, 376-391.