科技類股投資組合交易策略與決策研究

Tech-sector Stock Investment Portfolio Trading Strategy and Decision Making

林水金(Shui-Chin Lin) ； 陳啟斌(Chin-Bein Chen) ； 王光偉(Guang-Wei Wang) ； 董貴玲(Kuei-Ling Tung)

《創新研發學刊》 6卷1期 (2010/06) Pp. 60-79

https://doi.org/10.30080/JIRD.201006.0005

摘要

本研究先採用資料包絡分析法(DEA Porfolio Efficiency Index, DPEI)選取共同基金市場中之科技類股票型基金之績效最佳者（即績效值為1）；再評選各最佳基金之「前五名持股數」之總數做為本研究樣本，應用短期技術指標之灰關聯分析從中再篩選出排序為前五名之最佳績效的股票。最後，應用基因演算法與存貨模式求解各股票之最佳持有部位，進而建立最適持股數之投資組合。結果發現，本研究所提出之投資交易策略與決策模型能創造96年5月單月有10.92%及14.29%之月報酬率，此報酬率比目前市場上之科技類股票型基金平均報酬還要高。因此，本研究希冀能透過此一投資模式之建構，來提供投資者未來在選取投資組合及最適持股上之參考。

關鍵字

資料包絡分析法 ； 灰關聯分析 ； 基因演算法 ； 投資組合 ； 存貨管理

並列摘要

This study is based on the methods of data envelopment analysis (DEA), grey relational analysis and genetic algorithm to analyze the investment portfolio. The tech-sector mutual fund are chosen to implement the empirical study and decision marketing. First of all, applying the tech-sector mutual fund into the DEA in order to select the best performance of funds, which E(subscript k)=1 in Taiwan mutual fund market. Secondly, applying the grey relational analysis and short-term technique index into the total non-repeated stocks of the top five which are chose from the best performance of each fund, are applied to their order rank. These total stocks will be the samples for the study. Finally, the genetic algorithm is used to the inventory model to decide the holding position of the best five-ranked stocks.
The empirical evidence shows that the proposed investment trading strategy and decision making model gained 10.92% and 14.29% rate of return in May 2008, which was better than the all of tech-sector mutual fund for the average rate of return in the market. As the result, the model hopefully provides the best choice of investment portfolio and optimal holding position of stocks to the investors as an effective investing reference.

並列關鍵字

Data envelopment analysis ； grey relational analysis ； genetic algorithm ； investment portfolio ； optimal holding position