傳統財務學假設市場為效率市場，但現實市場中存在的異常現象，像是超額波動性(Excess Volatility)等現象並無法被合理解釋，因此學者轉而從心理學的角度去探討投資行為。展望理論說明人們實際在面對相同金額的獲利和損失時，損失的痛苦遠大於相同金額獲利所獲得的滿足，由於人們在面對損失時，恐慌與害怕的想法往往令人失去理智，因此市場上充斥著雜訊交易者(Noise trader)。雜訊交易者的錯估(Misperceptions)造成套利無法發揮預期的力量，雜訊因素是影響股價的重要原因之一。De Long et al.學者由理論的觀點提出DSSW模型，並假設錯估呈現常態分配，但根據展望理論再對應到股市中，投資者情緒處於悲觀或樂觀時所產生錯估的變異程度應該不同。因此本研究加入展望理論的概念，提出PT-DSSW模型，並結合遺傳演算法分成GA-DSSW及GA-PT-DSSW兩個系統，代入實際股價資料做驗證。結果顯示因雜訊交易者投資策略較短暫，因此採用訓練期間較短的雜訊交易模型具備較佳的預測能力，而PT-DSSW模型，將錯估定義為左右變異程度不相同的常態分配，在對於不同走勢的市場特性來說，本研究所提出的模型較具有解釋的能力。

Traditional financial theory assumes the market is efficient, but there are many phenomenons can’t be explained with “Efficient Market Hypothesis” reasonably, such as “Excess Volatility”, so many scholars attempt to discuss the investing behavior with psychology. “Prospect Theory” explains that when individuals face the same amount of money of gaining or losing, the distressed feel of losing will not be able to cover the satisfied feel of gaining. When individuals face the losing, they will be influenced by negative thinking, so there are many “noise trader” filling in the market. Noise trader’s “Misperceptions” causes the inefficiency of arbitrage, and the noise element is one of the important factors affecting the stock price. De Long et al.’s DSSW model assuming the misestimate appears normal distribution, but according to prospect theory, the emotions of investors which are optimistic or pessimistic should cause different levels of misestimates. In this article, we propose PT-DSSW model which combines the concept of prospect theory and genetic algorithm divides into two systems which are GA-DSSW and GA-PT-DSSW. From our experiments, because the periods of the noise trader’s investing strategies are short, the noise trade model of the shorter training period could be better forecasting ability, and PT-DSSW model defining misestimates as a normal distribution which of different variable levels between right and left tail has powerful explaining ability in different kinds of markets.

摘　要 iii
Abstract iv
誌 謝 vi
圖目錄 viii
表目錄 x
第一章 緒論 1
第二章 文獻回顧 4
2.1遺傳演算法(Genetic Algorithm, GA) 4
2.2展望理論(Prospect Theory) 8
2.2雜訊交易行為 10
第三章 研究設計 15
3.1研究架構 15
3.2遺傳演算法設計 16
3.2.1染色體編碼 17
3.2.2適應函數 17
3.3 PT-DSSW模型 18
第四章 實驗設計 24
4.1實驗一：針對不同走勢的市場特性做探討 25
4.2實驗二：觀測較適的訓練期間長短 33
4.3實驗三：比較GA-DSSW與GA-PT-DSSW整體的預測能力 36
第五章 系統介面 39
第六章 結論與建議 40
附錄一　PT-DSSW模型的細部公式推導 42
附錄二　關於實驗三於各期移動視窗的細部資料 46
參考文獻 52

[1]周賓鳳、池祥萱、周冠男與龔怡霖，行為財務學：文獻回顧與展望，證券市場發展季刊，14卷2期，2002，1-48。
[2]林豐澤，演化式計算下篇：基因演算法以及三種應用實例，智慧科技與應用統計學報，3卷，2005，29-56。
[3]賴素鈴、楊靜琪，台灣股市雜訊交易因素及其對股價影響性之研究－融合時間序列橫剖面迴歸模式，風險管理學報 ，6卷 1期，2004年3月，5-31。
[4]連立川、葉怡成、江宗原與楊豐銘，遺傳演算法建構台灣股市期貨買賣決策規則之研究，資管評論，14期，2006，47-61。
[5]陳隆麒、翁霓與郭敏華，雜訊交易對台灣地區投資人行為及股價之影響，證券市場發展季刊，7卷1期，1995年，101-123。
[6]薛立言與黃志傑，影響國內股市雜訊交易之因素分析，證券市場發展季刊，8卷3期，1996年，63-88。
[7]李春安，雜訊交易對貨幣政策與股價關聯性影響之研究，經濟論文叢刊，29卷3期，2001年，339-364。
[8]A. S. Kyle, “Continuous Auctions and Insider Trading”, Econometrica, Vol. 53, No. 6, Nov. 1985, pp.1315-1335.
[9]A. Shleifer and R. W. Vishny, “The Limits of Arbitrage”, Journal of Financial, Vol. 52, No. 1, Mar. 1997, pp. 35-55.
[10]A. Shleifer and L. H. Summers, “The Noise Trader Approach to Finance”, Journal of Economic Perspectives, Vol. 4, No. 2 , Spring 1990, pp. 19-33.
[11]D. Kahneman and A. Tveersky, “Prospect Theory: An analysis of Decision Under Risk”, Econometrica, Vol. 47, No. 2, Mar. 1979, pp. 263-291.
[12]E. F. Fama, “Efficient Capital Markets: A Review of Theory and Empirical Work”, Journal of Finance, Vol. 25, No. 2, May 1970, pp. 383-417.
[13]E. F. Fama, “The Behavior of Stock Market Prices”, Journal of Business, Vol. 38, No.1, Jan. 1965, pp. 34-105.
[14]F. Black, “Noise”, Journal of Finance, Vol. 41, No.3, Jul. 1986, pp.529-543.
[15]F. Allen, and R. Karlajainen, “Using Genetic Algorithms to find Technical Trading Rules”, Journal of Financial Economics, Vol. 51, No.2, Feb. 1999, pp. 245-271.
[16]F. Palomino, “Noise Trading in Small Markets”, Journal of Financial, Vol. 51, No. 4, Sep. 1996, pp. 1537-1550.
[17]I. Friend, and M. E. Blume, “The Demand for Risky Assets”, The American Economic Review, Vol.65, No.5, Dec. 1975, pp. 990-992.
[18]J. B. De Long, A. Shleifer, L. H. Summers and R. J. Waldmann, “The Size and Incidence of the Losses from Noise Trading”, Journal of Finance, Vol. 44, No. 3, Jul. 1989, pp. 681-696.
[19]J. B. De Long, A. Shleifer, L. H. Summers and R. J. Waldmann, “Noise Trader Risk In Financial Markets”, Journal of Political Economy, Vol. 98, No. 4, Aug. 1990 ,pp. 703-738.
[20]J. B. De Long, A. Shleifer, L. H. Summers and R. J. Waldmann, “The Survival of Noise Traders in Financial Markets”, Journal of Business, Vol. 64, No. 1, Jun. 1991 ,pp. 1-19.
[21]J. M. Poterba and L. H. Summers, “Mean Reversion in Stock Price: Evidence and Implications”, Journal of Finance Economy, Vol. 22, Oct. 1988, pp. 27-59.
[22]J. H. Holland, “Adaptation in Natural and Artificial System”, Ann Arbor: University of Michigan Press, USA, 1975.
[23]N.Barberis and R. Thaler, “A Survey of Behavioral Finance”, Handbook of Economics of Finance ,Chapter 18, Vol. 1, Part II, 2003, pp. 1053-1128.
[24]R. Gertner, ”Game Shows and Economic Behavior: Risk Taking on Card Sharks”, Quarterly Journal of Economics, Vol.108, No.2, May 1993, pp. 507-521.
[25]Rui Jiang and K. Y. Szeto, “Extraction of Investment Strategies based on Moving Averages: A Genetic Algorithm Approach”, IEEE 0-7803-7654-4, Mar. 2003, pp. 403-410.
[26]R. J. Shiller, “Stock Prices and Social Dynamics”, Brookings Papers on Economic Activity, Vol. 1984, No. 2, 1984, pp. 457-510.
[27]T. J. O’Brien, “A Simple and Flexible DCF Valuation Formula”, Journal of Applied Finance, Vol. 13, No.2, Winter 2003, pp. 54-62.