趨勢策略應用於台灣股票市場之研究

A Study of Investment Strategy Based on Trend at Taiwan Stock Market

周聖鈞(Sheng-Chun Chou)

指導教授 ： 吳忠敏

國立臺北科技大學/管理學院/工商管理研究所/博士(2013年)

摘要

台灣股票市場屬於淺碟型市場，容易受到外力的影響而產生劇烈的波動，再加上台灣股票市場以散戶投資人居多，具有股票持有期間短與週轉率高的特性，因此投資人一旦看錯趨勢方向往往會產生巨大的損失。本研究提出一個新的趨勢投資方法，首先透過股價趨勢模型將股價分為漲勢、盤整、跌勢等三種趨勢，克服過去僅能以股價或技術指標圖形訊息來判斷趨勢，造成過於主觀且無法量化的問題，接著建構出趨勢選股策略模型與趨勢投資策略模型。
趨勢選股策略模型利用股價趨勢模型所找出之趨勢，從個股過去的趨勢來決定是否將其選為投資組合。趨勢投資策略模型利用基因演算法的最佳化能力，在不同的股價趨勢中，產生不同的投資策略，協助投資人在不同趨勢下，找出適合的股票買賣時點。
本研究以台灣股票市場上市公司為研究對象，蒐集個股之股價日資料，藉此探討趨勢策略於台灣股票市場之投資績效，並與市場上常用的選股策略與投資策略做比較，以驗證趨勢策略之有效性。研究結果顯示，趨勢選股策略模型能夠藉由個股過去的趨勢來選擇出適合的投資組合，產生獲利。而趨勢投資策略模型在勝率與投資報酬率方面皆能產生良好的績效，進而在波動劇烈的台灣股票市場中，獲得報酬。

關鍵字

趨勢 ； 動能策略 ； 投資策略 ； 基因演算法

並列摘要

Most of investors use similar investment approaches in the stock market, no matter in bull period or in bear period. It may lead to unstable performance and relatively low profit. This study proposes a new trend based investment approach. First, the new price trend clustering model identifies three types of stock price movements: uptrends, sideways trends, and downtrends, in order to overcome limitations when diagnosing trend by referring to stock price, technical indicators and graphic messages only. Then we construct a trend based stock selection strategy model and investment strategy model based on trend.
Trend based stock selection strategy model is used for stock selection problem. We determine whether selected as investment targets through stock trend clustering model and then buy winners or buy losers. We also verify the effectiveness of trend based stock selection strategy model, which applied in the Taiwan stock market. Then we propose a investment decision model based on trend by using genetic algorithm to search for optimal solutions and the best investment strategies for different stock price trends.
This paper is based on daily stock trading records from the Taiwan stock market, which includes stock prices and trading volume. This paper has reviewed the performance of investment strategy based on the new trend based investment model. Also, common investment strategies have been compared with the new model, in order to proof its effectiveness. The results show that the trend based stock selection strategy model can choose a suitable portfolio and generate profit. Furthermore, the trend investment strategy model developed by using genetic algorithm performs better than other investment strategies. Research results prove the proposed new trend based investment model to be a stable and valid investment strategy.

並列關鍵字

Trend ； Momentum Strategy ； Investment Strategy ； Genetic Algorithm

參考文獻

1.王雍智、張澤與戴宏廩(2011)，風格投資-台灣股市的實證，東海管理評論，第13卷，第1期，1-46。

2.古永嘉與蔡垂君(2001)，介入模式應用於機構投資人鉅額交易訊息之研究，文大商管學報，第6卷，第1期，77-106。

3.李愷莉(2012)，與時變動之系統風險探討-以轉換迴歸模型為例，創新與管理，第9卷，第1期，55-88。

4.李顯儀與吳幸姬(2009)，技術分析資訊對共同基金從眾行為的影響，臺大管理論叢，第20卷，第1期，227-260。

7.林水金、陳啟斌、王光偉與董貴玲(2010)，科技類股投資組合交易策略與決策研究，創新研發學刊，第6卷，第1期，60-79。