近年來，很多研究發現過去的股票報酬能預測未來的股價，即投資者可利用過去的股價資訊， 發展出一套投資策略(Investment strategy) 以獲取異常報酬(Abnormal return)。這些投資策略可分成兩類： 第一類是逆勢策略(Contrarianstrategy)，即「買入過去的輸家(Loser)，賣出過去的贏家(Winner)」，該策略是依賴價格逆轉(Price reversal)。第二類是順勢策略(Momentum strategy)，即「買入過去的贏家，賣出過去的輸家」，這策略是根據價格延續(Price continuation)。本研究則是探討台灣股市從1978 年1月到2000 年6 月間是否出現價格逆轉與價格延續的現象。結果顯示產業順勢策略較個股順勢策略有利可圖。特別是，產業順勢投資策略（即買入過去的贏家產業組合，賣出過去的輸家產業組合）的高獲利從第1 個月一直延續至第36 個月，並且該策略在剔除上櫃公司與只考慮一半的子期間之下仍可獲利，顯示出該策略的穩定性。相反的，個股順勢策略(即買入過去的贏家股票，賣出過去的輸家股票)則是顯示出貧乏的績效與顯著較差的利潤。進一步，我們發現投資策略的高報酬不可能完全以風險補償和買賣偏誤來解釋。顯見尚有其他原因，這也許可能是來自於投資人對資訊反應不足、或是過度反應。
In recent years, many papers have documented that stock past returns can be used to predict stock future prices. Investors can develop some investment strategies based on past returns to achieve abnormal returns. These investment strategies can be grouped into two families---contrarian strategies (buying past losers and selling past winners) and momentum strategies (buying past winner and selling past losers). The contrarian strategy relies on price reversals and the momentum strategy bases on price continuations. This study is to explore whether exist price reversal and price continuation, on the Taiwan stock market form January 1978 to June 2000. Our results show that industry momentum strategies are more profitable than individual stock momentum strategies. Specifically, industry momentum investment strategies, which buy stocks from past winning industries and sell stocks from past losing industries, appear highly profitable over 1 to 36 months. Industry momentum strategies are robust to various specifications and methodologies, and they appear to be profitable even if the tests are applied to sample to disregard ROSE-listed stocks, and if only half of the sample period is considered. By contrast, individual stock momentum strategies, which buy past winning stocks and sell past losing stocks, are poor performance and significantly less profitable. Moreover, we find that high profitability of investment strategies cannot be fully explained by risk compensation and bid-ask bias. Obviously, there are also other reasons, possibly from investor overreaction and/or under-reaction to information.
