本篇論文試圖以行為財務學的視角建構台股週沖交易策略，考量到人類的
行為偏誤難以被完整的量化，且市場心理對於價量資訊間複雜的影響，因此本
研究採用結合極限梯度提升樹 (XGBOOST) 演算法結合行為財務特徵變數分別建
構預測未來五日報酬率之迴歸樹模型與預測未來五日漲跌機率的分類樹模型，
交叉比對結合成選股模型以應用到台灣股票市場。
在透過行為財務變數與 XGBoost 演算法模型所建構出的策略，於 2017 年至
2020 年 3 月底的報酬率可達到約 70%，年化報酬率 22%，並幾乎於全回測期間
內打贏大盤指數。

This paper attempts to construct a weekly stock trading strategy for Taiwan stocks from
the perspective of behavioral finance. Considering that behavior biases are too difficult to
be quantified and the complex impact of market psychology on price-volume information,
this paper uses eXtreme Gradient Boosting algorithm combines behavioral financial
characteristic variables to construct two models : a regression tree model that predicts the
future five-day return and a classification tree model that predicts the future five-day
probability of stock prices rising, to construct a stock selection model for application to the
Taiwan stock market.
The strategy constructed through behavioral financial variables and the eXtreme
Gradient Boosting algorithm model can achieve a return rate of about 70% from 2017 to the
end of March 2020, an annualized return rate of 22%. This performance beats market
benchmarks almost in the full period.

Content
論文審定書 i
摘要 ii
Abstract iii
Chapter 1. Introduction 1
Chapter 2. Literature Review 3
Chapter 3. Methodology 15
3.1 Experimental Structure 15
3.2 Data Collection and Features Construction 17
3.3 Model Features 20
3.4 Model Architecture 40
Chapter 4. Empirical Results 53
4.1 Model sensitivity analysis 54
4.2 Model Architecture Backtesting Performance 58
Chapter 5. Conclusion and Suggestions 63
5.1 Conclusion 63
5.2 Suggestion 64
Reference 65

Reference
Baker, Malcolm, and Jeffrey Wurgler (2006). Investor sentiment and the cross‐
section of stock returns. Journal of Finance 61, 1645–1680.
Ballings M., Poel D.V.d., Hespeels N., Gryp R. Evaluating multiple classifiers for
stock price direction prediction. Expert Systems with Applications, 42 (20) (2015), pp.
7046-7056
Barberis, N., Shleifer, A., Vishny, R. (1998). A model of investor sentiment.
Journal of Financial Economics 49, 307–343
Barberis, N., Shleifer, A., Vishny, R. (1998). A model of investor sentiment.
Journal of Financial Economics 49, 307–343.
Barberis, Mukherjee, Wang, 2016, N. Barberis, A. Mukherjee, Wang B. Prospect
theory and stock returns: an empirical test. Yale University, Hong Kong University of
Science and Technology and Fordham University (2016)
S. Basak, S. Kar, S. Saha, L. Khaidem, S. Roy Dey. Predicting the direction of
stock market prices using tree-based classifiers. The North American Journal of
Economics and Finance, 47 (2019), pp. 552-567
Bhardwaj, R.K. and L.D. Brooks (1992). The January Anomaly:Effects of Low
Share Price , Transaction Costs and Bid-Ask Bias. Journal of Finance, Vol. 47 ,1992,
PP.553-575.
Brad M. Barber, Terrance Odean (2000). Trading Is Hazardous to Your Wealth:
The Common Stock Investment Performance of Individual Investors. Journal of
66
Finance, 2000, vol. 55 (pg. 773-806).
Brad M. Barber, Terrance Odean (2001). Boys will be Boys: Gender,
Overconfidence, and Common Stock Investment. The Quarterly Journal of Economics,
Volume 116, Issue 1, February 2001, Pages 261–292.
Brown, Gregory W. and Michael T. Cliff (2004). Investor sentiment and the near‐
term stock market, Journal of Empirical Finance 11, 1– 27.
Chan, L.K.C. and J. Lakonishok , “Institutional Trades and Intraday Stock Price
Behavior,” Journal of Financial Economics, Vol.33, pp.173-199, 1993.
Chen, H., Chong, T. T. L., & Duan, X. (2010). A principal-component approach to
measuring investor sentiment. Quantitative Finance, 10(4), 339–347.
S.P. Chatzis, V. Siakoulis, A. Petropoulos, E. Stavroulakis, N. Vlachogiannakis.
Forecasting stock market crisis events using deep and statistical machine learning
techniques. Expert Systems with Applications, 112 (2018), pp. 353-371
Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In
Proceedings of the 22nd acm sigkdd international conference on knowledge discovery
and data mining (pp. 785794). ACM.
Chuang, Wei-I and Rauli Susmel (2011). Who is the More Overconfident Trader?
Individual vs. Institutional Investors. Journal of Banking and Finance, 35, 1626–1644.
R. Dash, P.K. Dash. A hybrid stock trading framework integrating technical
analysis with machine learning techniques. J Finance Data Sci, 2 (1) (2016), pp. 42-57
D.K. Ding, C. Charoenwong, R. Seetoh. Prospect theory, analyst forecasts, and
stock returns. Journal of Multinational Financial Management, 14 (2004), pp. 425-442
67
E.F. Fama (1970). Efficient capital markets: A review of theory and empirical work.
Journal of Finance, 25 (2) (1970), pp. 383-417
Eugene F. Fama (1970). Efficient Capital Markets: A Review of Theory and
Empirical Work. Journal of Finance (May 1970), pp. 383-417.
Friedman, J. H. (2001). Greedy function approximation: a gradient boosting
machine. Annals of statistics, 11891232.
George T. J. Hwang C. Y. 2004. The 52-week high and momentum investing.
Journal of Finance59:2145–76.
Hou Kewei, Peng Lin, Xiong Wei. A tail of two anomalies: The implications
of investor attention for price and earnings momentum,. 2006 Working Paper, Ohio
State University.
J. Li, J. Yu (2012). Investor attention, psychological anchors, and stock return
predictability. Journal of Financial Economics, 104 (2012), pp. 401-419.
T.K. Lee, J.H. Cho, D.S. Kwon, S.Y. Sohn. Global stock market investment
strategies based on financial network indicators using machine learning techniques.
Expert Systems with Applications, 117 (2019), pp. 228-242.
Kahneman, Daniel and Amos Tversky (1979). Prospect Theory: An Analysis of
Decision Under Risk. Econometrica, 47, 263–291.
Luo, Jin-Shuei and Chun-An Li (2008), Futures Market Sentiment and
Institutional Investor Behavior in the Spot Market: The Emerging Market in Taiwan,
Emerging Market Finance and Trade, 44, 70–86.
Odean, Terrance (1998). Are investors reluctant to realize their losses?, Journal
of Finance 53, 1775– 1798.
68
----, Michal Ann Strahilevitz and Brad M. Baber (2004), “Once Burned, Twice Shy:
How Naive Learning and Counterfactuals Affect the Repurchase of Stocks Previously
Sold,” Working paper, University of Arizona.
J. Patel, S. Shah, P. Thakkar, K. Kotecha. Predicting stock and stock price index
movement using Trend Deterministic Data Preparation and machine learning
techniques. Expert Systems with Applications, 42 (2015), pp. 259-268
Rahul Verma and Gokce Soydemir (2009). The impact of individual and
institutional investor sentiment on the market price of risk. The Quarterly Review of
Economics and Finance, 49 (3) (2009), pp. 1129-1145.
S. Basu (1977). Investment Performance of Common Stocks in Relation to Their
Price‐Earnings Ratios: A Test of the Efficient Market Hypothesis. The Journal of
Finance 3 (June 1977), 663– 82.
S. Gervais, T. Odean (2001). Learning to be overconfident. Review of Financial
Studies, 14 (2001), pp. 1-27
Maik Schmeling (2006). Institutional and individual sentiment: smart money and
noise trader risk? International Journal of Forecasting, 23 (2006), pp. 127-145.
S.I. Seok, H. Cho, D. Ryu (2019). Firm-specific investor sentiment and daily stock
returns. North American Journal of Economics and Finance .
Stambaugh, Robert F., Jianfeng Yu, and Yu Yuan (2012). The short of it: Investor
sentiment and anomalies. Journal of Financial Economics 104, 288– 302.
Tversky A., D. Kahneman (1974). Judgment under uncertainty: Heuristics and
biases. Science, 185 (1974), pp. 1124-1131.
Weber, M., & Camerer, C. F. (1998). The disposition effect in securities trading:
69
An experimental analysis. Journal Economic Behavior & Organization, 33(2), 167-184.
W. Zhang, W. Semmler. Prospect theory for stock markets: empirical evidence
with time-series data. J. Econ. Behav. Organ., 72 (3) (2009), pp. 835-849
X. Zhong, D. Enke. Forecasting daily stock market return using dimensionality
reduction. Expert Systems with Applications, 67 (2017), pp. 126-139.
F. Zhou, Q. Zhang, D. Sornette, and L. Jiang, “Cascading logistic regression onto
gradient boosted decision trees for forecasting and trading stock indices,” Applied Soft
Computing, vol. 84, p. 105747, 2019.