中文摘要
投資者最關心的問題莫過於能否賺到錢。台股報酬率的實證研究已有相當多的文獻得出相關的成果。對投資者而言，解釋變數能否對台股報酬率有好的預測效能顯然更為重要，所以本研究以1996年至2013年十八年期間共271家台灣一般產業上市公司資料為觀察值，以Hansen(1999)所提出的門檻迴歸模型(threshold regression model)探討台股上市公司之股票報酬率是否具有每股盈餘的門檻效果，在某個每股盈餘區間內及區間外，其總資產報酬成長率、每股營業利益、及淨值成長率對股票報酬率的影響方向與程度，再與傳統迴歸模型比較。
由實證結果可以發現，預測2012年的部分：傳統迴歸模型中，總資產報酬成長率、每股營業利益、每股盈餘皆具有顯著解釋能力，且為正相關；而門檻模型中以總資產報酬成長率及每股營業利益具有顯著且正相關的解釋能力。預測2013年的部分：傳統迴歸與門檻迴歸模型中，總資產報酬成長率及每股營業利益皆具有顯著且正相關的解釋能力。預測2014年的部分，傳統迴歸模型中，每股盈餘及淨值成長率有顯著且正向的解釋能力；門檻模型中則只有淨值成長率有顯著解釋能力且為正相關。

ABSTRACT
The major problem which investors the most care about is whether investment could make money or not. For investors, the explanatory variables whether can have a good prediction effectiveness on the returns on Taiwan stock is clearly even more important. In this study, the observed values were collected from 271 Taiwan firms from 1996 to 2013. A threshold regression model proposed by Hansen(1999) were used to test the threshold effect and to investigate whether earnings per share within a certain range, total assets growth rate, operating profit per share and net growth rate on the effect of Taiwan stock returns, and then the results were compared with the traditional regression model.
The empirical result shows total assets growth rate and operating profit per share are statistically significant and has a positive relationship with stock returns in both traditional regression and threshold regression in models of 2012 and 2013. Net growth rate is statistically significant and has a positive relationship with stock returns in both traditional regression and threshold regression in model of 2014. Net growth rate is statistically significant and has a positive relationship with stock returns in both traditional regression and threshold regression in model of 2014. Earnings per share (EPS) is statistically significant and has a positive relationship with stock returns only in traditional regression in models of 2012 and 2014.

參考文獻
1.李岱霙，2010，套利定價理論中之最適風險因子選取問題，國立高雄大學，碩士論文。
2.蕭山隆，2004，台股報酬多因子分析之研究：以外在變數及時間數列為例，朝陽科技大學，碩士論文。
3.利怡玟，2006，台灣上市公司經營績效與股票報酬率之關聯性，朝陽科技大學，碩士論文。
4.吳紹銘，2012，用門檻模型分析:經濟成長與匯率對ETF報酬的影響，國立中山大學，碩士論文。
5.李淑芬，2014，匯率失衡、出口與成長－非線性門檻模型分析，南華大學，碩士論文。
6.Dagenais, M.G.,1969, “A Threshold Regression Model” , Econometrica , Vol. 37, pp .193-203.
7.Fama, E.F., French, K.R. , 1995, “Size and book to market factors in earnings and returns” , Journal of Finance, Vol. 50, pp . 131–155.
8.Hansen, B.E., 1999, “Threshold effects in non-dynamic panels: Estimation, testing and inference”, Journal of Econometrics, Vol. 93, pp. 345-368.
9.Pari, R.A., Chen, S.N., 1984, “An Empirical Test of the Arbitrage Pricing Theory”, Journal of Financial Research, Vol. 7, pp. 121-130.
10.Ross, S. A., 1976, “The Arbitrage Theory of Capital Asset Pricing”, Journal of Economic, Vol. 13, pp. 341-360.