透過您的圖書館登入
IP:3.133.160.156
  • 學位論文

自組織映射圖神經網路應用於台灣股票型基金績效之研究

A Study on Self-Organizing Map Application for the Taiwanese Equity Fund Performance

指導教授 : 林秀怡
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


共同基金為目前市場上散戶投資人偏愛的投資工具之一,因其取得投資分析之資訊與資金較為有限,故股票型基金不失為一較佳的投資選擇,然而,在為數眾多的股票型基金中,如何選擇較績效持續較優的投資標的,實為一重大課題。有鑒於2008年金融海嘯引起全球經濟危機,全球股票市場遭受衝擊,台灣的股市也無法倖免,小眾投資人如何在此段異常期間,觀測出基金績效持續性的表現,對於投資人來說,亦為一重要關注議題。 本研究以台灣股票型基金對於在海嘯期間之績效衡量,研究期間以2005年至2011年,運用自組織映射圖神經網路模型將本研究之股票型基金採樣樣本進行無監督式分群,並針對分群後之多群群組,分別剖析探討其群性特質,並輔以海嘯期間各檔基金之投資效益加以分析解釋,藉此提供散戶投資人作為基金選購之投資依據。

並列摘要


The mutual fund is one of the favorite investment tolls for retail investors. Due to the limited information and funds, the equity fund is regarded as a better investment choice. However, it is a major issue to pick a well-performing investment target among many equity funds. In view of the global economic crisis caused by the financial tsunami in 2008, the global stock market was severely attacked. The stock market in Taiwan couldn’t be spared either. An important issue of concern for investors is to observe the continuous performance of funds during the abnormal period. This study tries to measure the performance of Taiwanese equity funds during the period of financial tsunami. The research period is between 2005 and 2011. The self-organizing map neural network model is applied to perform an unsupervised clustering on the sample of equity funds. This study also analyzes the clustering characteristics and explains the investment efficiency of each fund during the period of financial tsunami. The results can provide references for retail investors when choosing funds.

參考文獻


參考文獻
(1) 李純瑩,「國內股票型基金績效之影響因素研究」,世新大學經濟學系,碩士論文,民國九十二年。
(2) 李淑君,「以階層式自組織映射圖神經網路於動態金融環境之異常財務行為發現」,國立交通大學管理學院碩士在職專班資訊管理組,碩士論文,民國九十六年。
(3) 何杏華,「適應性聚類演算法應用於生物資訊」,國立成功大學資訊工程學系碩博士班,博士論文,民國九十六年。
(4) 邱憶如,「應用自組織映射圖神經網路及倒傳遞類神經網路於機構投資人融資券行為分析研究-以台灣50 期貨指數為例」國立交通大學資訊管理研究所,碩士論文,民國九十七年。

延伸閱讀