- 學位論文
智能選股下外資、自營商和主力在個股之投資績效分析
Foreign Direct Investors, Dealers, and Market Makers’ Stock Picking Algorithms and Profitability Analysis
摘要
過去學者和投資人於金融市場中常針對法人在股票及期貨市場進出做了許多研究，也對法人在股票市場和匯率連動的影響多有著墨。對此，甚少將法人細分出外資、投信和自營商來個別探討，甚至只提到外資單一身分，而在台灣市場除了法人的身分外，也有著不為少數的散戶和較少人注意到的主力，它們也對股票市場有著舉足輕重的影響力，對此本研究將法人細分取其中外資、自營商及常人較陌生的主力作為本研究的標的，並以智能選股系統來蒐集三者分別的投資績效，透過交叉比對後，看其能否在瞬息萬變的資本市場中脫穎而出。 本研究主要是透過群益證劵的「智能選股」系統，來追蹤跟蒐集外資、自營商和主力每週買超個股標的的前20名。經由連續6個月資料蒐集並使用T檢定和ANOVA來比較外資、自營商和主力三者之投資績效優劣，並將之提供給投資者做為日後投資策略的工具。 透過統計結果得到外資、自營商、主力多數績效優於大盤績效，將其以T檢定得到主力多半績效是優於外資和自營商，並同時發現只要加入主力比較，其統計結果會有顯著差異，將三者納入大盤績效一併以ANOVA分析在統計上也得到顯著差異故，其統計上顯著差異原因來自主力表現最為突兀，且主力變異數大於另外兩者，故呈現顯著性差異。
並列摘要
Over the past decades, scholars have done numerous studies on institutional investor’s investing strategy, and their impact on the stock market. However, different groups of institutional investors, which included foreign direct investors, stock dealers, and stock traders, are rarely differenciate during studies. Furthermore, Taiwan’s stock market also comprised of a small group of market makers, whom are market participants that have significant influence on the stock market. This study will focus on the 3 major investor groups, which are foreign direct investors, stock dealers, and market makers. The data will be collected throught the stock selection system, to analyze their respective investment performance, and using cross-matching techniques to explain which group of investers can outperform in the ever-changing capital market. This study analysis the stock picking algorithms for foreign direct investors, stock dealers, and market makers, select their top 20 stocks purchased on weekly basis for a 6 month period, using T-test and ANOVA to analyze whether the stock picking strategy and performance, and provide suggestions for investors to make investment decisions. The statistical findings imply that stock trader outperform the foreign investor and stock dealer most of the time. Meanwhile, there is a statistically significant difference in the investment performance as comparing with the market. The statistically significant difference derives from stock dealer’s outstanding performance, and variance is greater than the other data sets, which lead to significant difference.