Responsive image
博碩士論文 etd-0703116-130605 詳細資訊
Title page for etd-0703116-130605
論文名稱
Title
以情境吃角子老虎機演算法推薦股票投資行為的研究
Using Contextual Multi-Armed Bandit Algorithms for Recommending Investment in Stock Market
系所名稱
Department
畢業學年期
Year, semester
語文別
Language
學位類別
Degree
頁數
Number of pages
59
研究生
Author
指導教授
Advisor
召集委員
Convenor
口試委員
Advisory Committee
口試日期
Date of Exam
2016-07-07
繳交日期
Date of Submission
2016-08-03
關鍵字
Keywords
情境式多拉桿拉霸問題、個人化推薦系統、股票推薦系統、情境式拉霸問題、線性上信賴界
LinUCB, Contextual Bandit Problem, Stock Recommendation, Contextual Multi-Armed Bandit, Personalized Recommendation System
統計
Statistics
本論文已被瀏覽 6100 次,被下載 295
The thesis/dissertation has been browsed 6100 times, has been downloaded 295 times.
中文摘要
情境式拉霸問題 (Contextual Bandit Problem) 經常被使用來模擬線上推薦的應用,像是文章、音樂、影片等推薦系統。線性上信賴界(LinUCB)是目前解決情境式拉霸問題的演算法之一,它主要使用線性回歸並且從環境當中所得到的回饋(feedback)進行不斷的學習並更新其內部的模型。然而我們觀察到在股票投資市場當中,使用情境式拉霸問題來解決股票推薦問題的應用少之又少,大部分研究推薦目的為投資營利,並非根據投資者本身的投資風險屬性、投資標的的特性推薦他們符合投資屬性的股票。
我們提出一個情境式拉霸問題模型來模擬推薦股票給使用者的個人化推薦系統。情境式多拉桿拉霸問題模型從投資者過往的投資紀錄找出他的投資屬性,再根據這些屬性來推薦股票的組合。而股票組合是從公司財務分析的基本面及股票變化的技術面二者分類出來的結果,決定推薦組合後,再根據推薦組合和所有股票的相似性去做排名,然後推薦股票。
我們實證資料來源是網路上的模擬投資股市的資料集,實驗的結果顯示我們提出的方法在推薦股票的領域比現有的方法好。
Abstract
The Contextual Bandit Problem (CMAB) is usually used to recommend for online applications on article, music, movie, etc. One leading algorithm for contextual bandit is the LinUCB algorithm, which updates internal linear regression models by the partial feedback from the environment. However, we observe that CMAB is rarely used in the stock recommendation, while most of the recommendations are for the purpose of profit, and ignore investor’s features (risk tolerance, investment features, and the others).
We propose a personalized recommendation system for stock by using contextual multi-armed bandit algorithm. We take investor’s investment records as user features, and recommend the “arm”, which is a type of stock, based on two kinds of analysis, the technical and fundamental analysis. To the chosen arm, we rank the stocks according to the similarity of the stock and the arm. Our experiment is base on an online investment dataset, and the result demonstrates that our method outperforms other algorithms.
Our experiment dataset collects simulation investment on the online website, and the result demonstrates that our method outperforms other algorithms.
目次 Table of Contents
TABLE OF CONTENTS
CHAPTER 1 – Introduction 1
1.1 Background 1
1.2 Motivation 3
CHAPTER 2 – Literature Review 6
2.1 Definition of Multi-armed bandit Problem 6
2.2 Bandit Algorithms Overview 7
2.2.1 ϵ - greedy 7
2.2.2 Upper Confidence Bounds (UCB) 8
2.3 Contextual Multi-armed bandit 9
2.3.1 LinUCB 10
2.3.2 Application in CMAB 13
2.4 Investment Measure 14
2.4.1 Sharpe Ratio 14
2.4.2 Prospect Theory 15
2.4.3 Stock Measurement 16
CHAPTER 3 – Our Approach 19
3.1 Architecture of our Approach 19
3.2 Definition of Arms 20
3.3 The LinUCB Algorithm 22
3.4 The Stock Recommendation 25
CHAPTER 4 – Data Description 31
4.1 Data Collection 31
4.2 Data Description 33
CHAPTER 5 – Evaluation 35
5.1 Experiment Design 35
5.1.1 Compared Algorithms 35
5.2 Performance Measure 37
5.2.1 Cumulative Reward (Rank Ratio) 37
5.2.2 Cumulative Reward (Sliding Window) 37
5.3 Experiment Results 38
5.3.1 Parameter Tuning 38
5.3.2 Comparison of Features 41
5.3.3 Size of the Data 42
CHAPTER 6 – Conclusion 46
Reference 48
參考文獻 References
Reference
[1] Pazzani, M. J., & Billsus, D. (2007). Content-based recommendation systems. In The adaptive web (pp. 325-341). Springer Berlin Heidelberg.
[2] Sarwar, B., Karypis, G., Konstan, J., & Riedl, J. (2001, April). Item-based collaborative filtering recommendation algorithms. In Proceedings of the 10th international conference on World Wide Web (pp. 285-295). ACM.
[3] Schafer, J. B., Konstan, J. A., & Riedl, J. (2001). E-commerce recommendation applications. In Applications of Data Mining to Electronic Commerce (pp. 115-153). Springer US.
[4] Lynch, C. (2001, June). Personalization and recommender systems in the larger context: New directions and research questions. In Second DELOS Network of Excellence Workshop on Personalisation and Recommender Systems in Digital Libraries.
[5]Linden, G., Smith, B., & York, J. (2003). Amazon. com recommendations: Item-to-item collaborative filtering. IEEE Internet computing, 7(1), 76-80.
[6] Miller, B. N., Albert, I., Lam, S. K., Konstan, J. A., & Riedl, J. (2003, January). MovieLens unplugged: experiences with an occasionally connected recommender system. In Proceedings of the 8th international conference on Intelligent user interfaces (pp. 263-266). ACM.
[7] Li, L., Chu, W., Langford, J., & Schapire, R. E. (2010, April). A contextual-bandit approach to personalized news article recommendation. In Proceedings of the 19th international conference on World wide web (pp. 661-670). ACM.
[8] Hariri, N., Mobasher, B., & Burke, R. (2014, October). Context adaptation in interactive recommender systems. In Proceedings of the 8th ACM Conference on Recommender systems (pp. 41-48). ACM.
[9] Kuleshov, V., & Precup, D. (2014). Algorithms for multi-armed bandit problems.arXiv preprint arXiv:1402.6028.
[10] Watkins, C. J. C. H. (1989). Learning from delayed rewards (Doctoral dissertation, University of Cambridge).
[11] Auer, P., Cesa-Bianchi, N., & Fischer, P. (2002). Finite-time analysis of the multiarmed bandit problem. Machine learning, 47(2-3), 235-256.
[12] Walsh, T. J., Szita, I., Diuk, C., & Littman, M. L. (2009, June). Exploring compact reinforcement-learning representations with linear regression. In Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence (pp. 591-598). AUAI Press.
[13] Chu, W., Li, L., Reyzin, L., & Schapire, R. E. (2011). Contextual bandits with linear payoff functions. In International Conference on Artificial Intelligence and Statistics (pp. 208-214).
[14] Bouneffouf, D., Bouzeghoub, A., & Gançarski, A. L. (2012). Hybrid-ε-greedy for mobile context-aware recommender system. In Advances in Knowledge Discovery and Data Mining (pp. 468-479). Springer Berlin Heidelberg.
[15] Wang, X., Wang, Y., Hsu, D., & Wang, Y. (2014). Exploration in interactive personalized music recommendation: a reinforcement learning approach. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), 11(1), 7.
[16] Shen, W., Wang, J., Jiang, Y. G., & Zha, H. (2015, June). Portfolio choices with orthogonal bandit learning. In Proceedings of the 24th International Conference on Artificial Intelligence (pp. 974-980). AAAI Press.
[17] Nagy, R. A., & Obenberger, R. W. (1994). Factors influencing individual investor behavior. Financial Analysts Journal, 50(4), 63-68.
[18] Markowitz, H. (1952). Portfolio selection. The journal of finance, 7(1), 77-91.
[19] Sharpe, W. F. (1994). The sharpe ratio. The journal of portfolio management, 21(1), 49-58.
[20] Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica: Journal of the econometric society, 263-291.
[21] Barberis, N., Huang, M., & Santos, T. (1999). Prospect theory and asset prices (No. w7220). National bureau of economic research.
[22] Hsu, Y. L., & Chow, E. H. (2010, August). House money and investment risk taking. In 23rd Australasian Finance and Banking Conference.
[23] Berndt, D. J., & Clifford, J. (1994, July). Using Dynamic Time Warping to Find Patterns in Time Series. In KDD workshop (Vol. 10, No. 16, pp. 359-370).
[24] Wang, X., Mueen, A., Ding, H., Trajcevski, G., Scheuermann, P., & Keogh, E. (2013). Experimental comparison of representation methods and distance measures for time series data. Data Mining and Knowledge Discovery, 26(2), 275-309.
[25] Chung, F. L., Fu, T. C., Ng, V., & Luk, R. W. (2004). An evolutionary approach to pattern-based time series segmentation. Evolutionary Computation, IEEE Transactions on, 8(5), 471-489.
[26] Müller, M. (2007). Information retrieval for music and motion (Vol. 2). Berlin: Springer.
電子全文 Fulltext
本電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
論文使用權限 Thesis access permission:自定論文開放時間 user define
開放時間 Available:
校內 Campus: 已公開 available
校外 Off-campus: 已公開 available


紙本論文 Printed copies
紙本論文的公開資訊在102學年度以後相對較為完整。如果需要查詢101學年度以前的紙本論文公開資訊,請聯繫圖資處紙本論文服務櫃台。如有不便之處敬請見諒。
開放時間 available 已公開 available

QR Code