利用灰色理論優化投資策略

Abstract

本研究建立在技術分析的有效性之下，將技術分析之策略以程式的方式表達，並利用回測的方式以找到更具有獲利能力之投資方式。策略所使用的指標有移動平均線、相對強弱指標、動量…等。台灣指數期貨市場可能屬於一個資訊不對稱的市場，於是本研究將採取兩個方法進行探討。第一個方式是利用能夠處理不明確資訊的灰色關聯度分析方法，而此方法能使投資策略所選之指標能夠更貼近市場實際走勢。第二個方式是利用能夠將看似毫不相關之事物，而從中找出它們相似的特性之灰色統計聚類方法，以增加策略本身的獲利能力，並降低策略本身之風險。 本研究結果顯示: 1.灰關聯參數指標相較於一般投資大眾所使用之指標參數，擁有更大的獲利能力和低風險特性。 2.灰色統計聚類能有效的找出台灣指數期貨相對低點之指標特性，而無法找出相對高點之指標特性。而加入灰色統計聚類出場條件之策略，獲利能力明顯地提升，風險明顯地下降，勝率明顯地提高。

This research is based on the fact that technical analysis method is useful, and change the strategy of technical analysis into the program, and use the backtesting method to find out a more profitable way of investment. The indicators used to compose the strategies are the Moving Average Lines and Relative Strength Indicator and Momentum Indicator...and so on. A asymmetric information may exist in the Taiwan Stock Index Future market, hence the research will be dealt in two methods. The first method is to combine the indicators with the Grey Relation Analytical Method, and the advantage of this method is that the trend between the indicators and the Taiwan Stock Index Future market will more similar. The second method is to use Grey Cluster Method to find out the characteristics of the indicators when the Taiwan Stock Index Future market is at the relative high or relative low in order to enhance the profitable ability and lower the risk of strategies. The results are as follows: 1.The indicators combined with the Grey Relation Analytical Method is more profitable and lower risk in comparison with the original indicators. 2.Using Grey Cluster method is effective to find out the characteristics of the indicators when the market is at relative low, but not effective for finding out the relative high. 3.The strategies which the clearing position ways combined with Grey Cluster method is joined are more profitable、lower risk and higher winning ratio.