本研究以技術指標乖離率為基礎的配對交易策略,來檢驗效率市場的有限性。Shleifer and Vishny (1997)認為證券市場的價格偏離(prices diverge)無法立即回歸正常價格,原因為在業績導向下的套利可能不完全有效使價格偏離回歸正常,且噪音現象使套利需要承受更多風險,因此產生套利的有限性;Cai, Faff and Shin (2012)延續Shleifer and Vishny 之研究,透過非線性模型證實,當價格偏離越大時,套利將越快發生,因此偏離也將越快消失。本研究以乖離率作為辨識股價偏離依據,挑選出價格偏離較顯著之股票驗證Cai et al. 之理論。本研究選取台灣證券交易所2007 年至2015 年之上市公司作為樣本,以五日乖離率挑選出正負乖離之股票來採取不同配對交易策略。檢視發生價格偏離之股票,適用於動能策略抑是逆勢策略。實證結果發現,逆勢策略優於動能策略,逆勢策略無論配對幾家股票皆有正向報酬,且配對5 筆、20 筆、34 筆及69 筆股票分別在不同持有期間下,以配對5 筆在持有半年之後,報酬最高。逆勢策略主要是透過價格反轉來獲取報酬,本研究發現選取越極端價格偏離的股票,來執行逆勢策略越能獲取報酬,由此證實Cai, Faff and Shin 看法,也就是投資人錯誤認知(misperceptions)會引發所謂的套利受限。
This study tests the limits of arbitrage by the pair trading with BIAS, a well-known technical indicator. Shleifer and Vishny (1997) arguments that the price diverge in the stock market can not revert to the normal level in time because of the performance-driven arbitrage and the existence of noise trading, then results in limited arbitrage. This study tries to verify this argument. This study uses BIAS as the proxy of price divergence, screens out significantly divergent stocks to run empirical tests. The results show that contrarian strategy is better than momentum one, the more divergent, the better the portfolio is. This finding coincides with the argument of Shleifer and Vishny (1997), i.e. persistence of 'mispricing deepen'