On the Two-Stage Estimation of the Fama-French Three Factor Model: Evidence from Taiwan

Abstract

關於因子模型的參數估計問題，文獻中常以兩階段迴歸處理。本文以台灣股票市場為例，檢驗一些常用的兩階段估計方法在Fama-French三因子模型參數估計的影響。證據大多顯示支持市場因子，而支持高－低淨值市價比的證據微弱。風險貼水的顯著程度取決於分組與逐年更新的程序；誤差變項與非同步交易的調整並對結果不會造成太大的影響。

In this paper, a few common approaches to implementing the two-stage test on the Fama-French three-factor models are examined for Taiwan's stock market. Much of the evidence favors the MKT factor, while the support for the HML factor is weak. The significance of the risk premium is deeply dependent on the grouping and rolling procedure. Adjustments for errors-in-variable or non-synchronous trading do not affect the results very much.