行為投資理論建構下投資組合之研究

Portfolio Selection Applying Behavioral Portfolio Theory

楊千億(Michael-Nayat Young)

指導教授 ： 張國華

中原大學/工學院/工業工程研究所/碩士(2014年)

https://doi.org/10.6840/cycu201400037

摘要

描述投資人在投資上有別於一般假設的投資行為的研究稱之為行為投資學(behavior portfolio theory, BPT)。 行為財務自提出展望理論(prospect theory)的經濟學家Kahneman於2002年獲得諾貝爾經濟學獎之後已逐漸獲得重視。在當今競爭激烈且強調客製化服務的金融市場，研究行為投資學下的投資組合最佳化也更顯重要， 再加上電腦計算工具的功能性與使用方便性的快速提高， 也使得行為投資學下的投資組合最佳化更加可行。
本論文的目的是考量投資人的投資行為與投資心理層面並藉由將它們量化以建構更貼切投資人屬性個別最佳化的投資組合。 本論文探討模式以分三階段量化上述投資的行為: 我們量化投資人不理性的行為以及其對投資標的物的影響; 我們以個人主觀機率分布量化投資人對未來景氣的評估; 我們以數理規劃模式量化不同心理帳戶之下最佳化個別的投資組合。在過去的研究中並未有將三階段整合一起討論的研究。
我們以線性迴歸分析投資人不理性的行為對投資標的物的關係，並以迴歸分析式子模擬產生未來投資標的物的報酬率; 我們以SP/A 理論量化個人主觀機率分布; 我們將心理帳戶分為風險愛好心理帳戶、一般風險心理帳戶以及風險規避心理帳戶。由測試的結果得知使用考慮不理性行為模擬資料的投資組合會較使用原始歷史資料的投資組合會有較好的表現，也會更貼切各個心理帳戶對風險的要求。在另一方面，測試結果一般而言也較大盤表現的好。
現在由於資訊與投資產品多元下，投資者會更有機會找到貼切自己個人需要的投資組合。行為投資學理論可幫助投資人能夠更精確的找到所需要的投資組合。本研究的結果將可提供投資者在建構個人投資組合時有可行的參考架構。

關鍵字

行為投資學 ； 心理帳戶 ； SP/A理論 ； 展望理論 ； 投資組合最佳化

並列摘要

In this study, we determine the market period using the Relative Strength Index and tested the top 150 securities in the MSCI Taiwan Index for the significant effects of irrational behaviors during Bull, Bear, and Normal Market Periods. From the top 150 securities, we narrow down our investment objectives to those securities showing effects of at least 1 irrational behavior. Considering the significant irrational behavior indexes, we use linear regression equations to generate 20000 scenarios for the investment objectives, then through SP/A theory and our developed estimator for fear and hope levels we assign probabilities to the possible return scenarios of the securities and the market. We then divide our investment into 3 mental accounts (MA-1, MA-2, and MA-3) which respectively have risk-seeking, risk neutrality, and risk aversion attitudes towards different goals. We used the inverse of the safety-first model, mean-variance model, and safety first model respectively for each mental account to calculate our portfolios. We use the generated scenarios and the historical data and applied mixed-integer programming by means of the AIMMS solver to test our models. We call the portfolios using historical data as RS-HD and SF-HD for the risk-seeking and safety-first mental account respectively. In the end, our portfolios outperformed the market as well as the portfolios using the historical data. Through the 100 weeks of test period, although RS-HD have more instances of returns at least 5% but overall our MA-1 account outperformed their RS-HD counterparts by winning 80% of their total comparison; our MA-3 portfolios also outperformed their SF-HD counterparts in terms of satisfying the threshold limit of having returns less than -5% at 5%, using our MA-3 portfolio selection model we only had 2 instances of returns less than -5% while the SF-HD had 6 instances. Our MA-3 portfolios also won 81.82 of the total comparison with SF-HD. Finally, our MA-1 and MA-3 portfolios were also able to outperform the market in terms of mean return and cumulative returns.

並列關鍵字

prospect theory ； SP/A theory ； mental account ； Behavior portfolio theory ； portfolio optimization

參考文獻

Baptista, A. M. (2012). Portfolio selection with mental accounts and background risk. Journal of Banking & Finance, 36(4), 968-980. doi: 10.1016/j.jbankfin.2011.10.015

Barber, B. M., & Odean, T. (2000). Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors. The Journal of Finance, 55(2), 773-806. doi: 10.1111/0022-1082.00226

Barber, B. M., & Odean, T. (2001). Boys will be Boys: Gender, Overconfidence, and Common Stock Investment. The Quarterly Journal of Economics, 116(1), 261-292. doi: 10.1162/003355301556400

Basu, S. (1983). The relationship between earnings' yield, market value and return for NYSE common stocks. Journal of Financial Economics, 12(1), 129-156. doi: 10.1016/0304-405x(83)90031-4

Conrad, J., & Kaul, G. (1998). An anatomy of trading strategies. Review of Financial Studies, 11(3), 489-519. doi: 10.1093/rfs/11.3.489