黃志典臺灣大學:國際企業學研究所劉宜達Liu, Yi-DarYi-DarLiu2010-05-112018-06-292010-05-112018-06-292008U0001-1807200822372600http://ntur.lib.ntu.edu.tw//handle/246246/182605本論文探討兩種合成投資策略：(1)買高益本比股票、賣低益本比股票，並且買高價格動能股票、賣低價格動能股票的「混合型投資策略」；(2)買高益本比、高價格動能股票，並且賣低益本比、低價格動能股票的「掺雜型投資策略」。本論文用兩種方法比較合成投資策略和原本的益本比投資組合(買高益本比股票、賣低益本比股票)、價格動能投資組合(買高價格動能股票、賣低價格動能股票)的績效：(1)比較平均異常報酬的t值；(2)比較異常報酬的平均值和標準差的信賴區間。本論文的樣本期間為1998年1月至2007年12月，樣本股票為曾在樣本期間上市的普通股。結果發現：(1)不管是用那一種方法，合成投資策略皆可能優於原本的投資組合；(2)儘管限制放空並扣除交易成本，合成投資策略仍然有可能打敗大盤。這些結果顯示，結合兩種負相關的選股指標可顯著地提昇投資績效。此外，本論文亦更新價值效應和動能效應在台灣股市的證據，並討論到價值因子和動能因子之間的關係。This thesis examines two composite strategies: (1) the strategy of buying high earnings-to-price (EP) stocks, selling low EP stocks, buying high price momentum (PM) stocks, and selling low PM stocks, or the “mix strategy;” (2) the strategy of buying high EP, high PM stocks and selling low EP, low PM stocks, or the “blend strategy.” Two methods are used to compare the composite strategies with the original EP portfolio (buying high EP stocks and selling low EP stocks) and PM portfolio (buying high PM stocks and selling low PM stocks): (1) comparing average abnormal returns’ t-values; (2) comparing confidence intervals for abnormal returns’ means and standard deviations. Using the Taiwan stock market data from January 1998 to December 2007, we find that: (1) the composite strategies could outperform the original portfolios, regardless of the comparison method; (2) even though we impose short selling restrictions and high transaction costs on them, they could still beat the market. These results show that combining two negatively correlated stock selection indicators can significantly improve investment performance. Updated evidence on value and momentum effects and relations between value and momentum factors in the Taiwan stock market are also documented.Table of Contentshesis Defense Committee Signature Form (Chinese)…………………………….......i cknowledgements (Chinese)………………………………………………………...iibstract (Chinese)……………………………………………………………………iiibstract………….……………………………………………………………………ivable of Contents……………………………………………………………………...vist of Figures………………………………………………………………………..viiist of Tables………………………………………………………………………...viii. Introduction 1. Literature reviews 1.1. Review of literature on value effects 2.2. Review of literature on momentum effects 3.3. Review of literature on the interaction between value and momentum effects 5. Hypotheses 5.1. Definition of mixing and blending 5.2. Relation between the earnings-to-price ratio and price momentum 6. Data 7.1. Variable definition 7.2. Sample 8.3. Performance of the Taiwan stock market during the sample period 9. Methodology 11.1. Test of H1: Factor portfolios’ abnormal returns 11.2. Test of H2: Correlation between two factors 12.3. Test of H3: Correlation between two factor portfolios’ abnormal returns 13.4. Test of H4: Composite strategies’ abnormal returns 13.5. Further remarks: Methodology justification 15. Empirical results 16.1. Test of H1: Factor portfolios’ abnormal returns 17.2. Test of H2: Correlation between two factors 20.3. Test of H3: Correlation between two factor portfolios’ abnormal returns 22.4. Test of H4: Composite strategies’ abnormal returns 27. Conclusions and implications 34.1. Conclusions 34.2. Applications 35ppendices 36.1. Evidence on value and momentum effects in the Taiwan stock market 36.2. Announcement deadlines of quarterly financial data in Taiwan 37.3. Sample size at each period 38.4. Number of stocks in a quintile at each period 38.5. Comparison of this study and past studies of the earnings-to-price ratio and price momentum 39eferences 40ist of Figuresig. 1: Time-series plot of market returns and risk-free returns 11ig. 2: Illustration of the relation between the EP and PM portfolios’ abnormal returns 26ig. 3: Confidence boxes of the abnormal returns of the EP portfolio, the PM portfolio, mix (0.75,0.25), and blend (0.75,0.25) (1998–2007) 30ig. 4: Time-series plot of the cumulative wealth of the bank deposit, Taiex, long-only mix (0.75,0.25), and long-only blend (0.75,0.25) 33ist of Tablesable 1: Market returns and risk-free returns during the sample period 10able 2: Performance of portfolios based on factors 18able 3: Correlation coefficient between two factors’ rankings 21able 4: Relation between two factor portfolios’ abnormal returns 23able 5: Comparison of original portfolios and composite strategies’ performance 28able 6: Cumulative wealth and geometric average returns of the bank deposit, Taiex, long-only mix (0.75,0.25), and long-only blend (0.75,0.25) 32application/pdf715245 bytesapplication/pdfen-US混合型投資策略掺雜型投資策略益本比價格動能mixingblendingearnings-to-price ratioprice momentumthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/182605/1/ntu-97-R95724003-1.pdf