儘管多角化投資對於消除特殊風險非常重要，但仔細研究個人投資者或稱散戶投資人的投資組合，可以發現這些投資者通常並未實現多角化投資，Bali, Nurset & Robert(2014)論文中，引入了一種新的股票報酬尾端共變異數風險的混合指標，其動機是個人投資者持有的投資組合多角化不足，並研究其橫截面預測能力，在台灣，散戶投資者佔比比美國來得高，因次我們想要建立一個類似的尾端混和風險指標，來測試這種混和尾端風險是否在台灣市場依然有效，這種共變異數是通過單個股票報酬分佈的左尾狀態來衡量的，而不是像標準系統風險衡量指標以市場報酬為評判標準。我們發現混合尾端風險（H-TCR）與預期股票報酬之間存在正相關關係，並且年化報酬大約為10%，這比美國股市的報酬來得高，而個股獨有風險或系統尾端風險變數的結果則無顯著或甚至是負面的。最後我們建構迴歸模型，並進行多個穩健性測試，其結果依然呈現混和尾端風險能夠產生超額報酬。;Though diversified investments are important to eliminate specific risks, a careful study of individual investors or portfolios of retail investors can be found that these investors often do not achieve diversified investments. Bali, Nurset & Robert (2014) paper, introduced a new hybrid indicator of the risk of the total variance of stock returns at the end of the stock. The motivation is that the portfolio of individual investors holds the diversification of the portfolio and studies its cross-sectional prediction ability. In Taiwan, the proportion of retail investors is higher than that of the United States, so we want to establish a similar hybrid tail risk indicator to test whether the risk of this hybrid tail is still valid in the Taiwan market. This co-variation is measured by the left-tail state of the distribution of individual stock returns. Instead of standard market risk measures, market rewards are used as criteria. We find a positive correlation between hybrid tail covariance risk (H-TCR) and expected stock returns, and annualized compensation is about 10%, which is higher than the compensation of the US stock market, while individual stocks have unique risks or system tail risks. The results of the variables are not significant or even negative. Finally, we construct a regression model and conduct multiple robustness tests. The results still show that the hybrid tail risk of can generate excessive rewards.
