影響機構投資人在臺灣指數股票型基金(ETFs)的投資決策因素?

Abstract

本論文透過馬可夫轉換模型區分不同的財務市場景氣狀態，並將之作為解釋變數，同時加入其他外生解釋變數，以三大法人買賣超ETF股數作為被解釋變數並採用向量自我迴歸模型(VAR模型)，目的在於:(1)檢視三大法人於台灣ETF市場的投資行為，是否受到不同的財務市場景氣狀態及市場投資人情緒影響，而投資於不同類型的ETF?(2)三大法人是否有彼此影響之情況? 希望可以幫助投信公司了解三大法人於臺灣ETF市場的動向，並做出相對的措施因應。

The purpose of this study is to investigate the factors which drive institutional investors’ investment decisions in Taiwan ETFs market. The thesis first predicts phases of the financial cycle in a Markov switching framework. VIX index is used to identify the turning points for a transition to a volatile state. The state variable, other institutional investors’ investment behavior in ETFs, and investor's sentiment variables are then use to explain the behaviors of three major institutional investors’ ETF shares in a vector autoregressive (VAR) framework. In particular, the study asks the following two research questions: (1) Whether institutional investors will invest in different types of ETFs in various financial market states, and (2) Whether there will be any interactions among the three types of major institutional investors in their ETF investment. Understanding the institutional investor behavior can help investment trust companies to promote ETFs in various financial states.