3.2. Three-factor alphas of sorted portfolios
Following a few existing studies on international stock returns (e.g., Fama and French, 1998; Griffin, 2002;
Hou et al., 2006), we use the Fama and French (1993) three-factor model to compute alphas for decile
portfolios in each local market:
Ritj−RFtj = αij + bijRMRFtj + siSMBtj + hijHMLtj + εitj ð2Þ
where Ritj−RFtj is the monthly returns of decile portfolio I in month t in excess of the monthly risk free
rate for market j. RMRFtj is the value-weighted market portfolio (including all stocks in each individual
market) return in excess of the risk free rate, and SMBtj and HMLtj are the monthly size and book-tomarket
factors for market j.
7 We follow the procedures described in Fama and French (1993) to
construct the factors for each Asian market, with the details provided in Appendix B.
The results are in Panel C of Table 2. Except for Taiwan (with insignificantly negative alpha), the three factor
alphas for the D10–D1 spreads are significantly negative in the rest of Asianmarkets, largely consistent with the
pattern on the monthly return spreads.
In the last two columns of the table, we report the F-statistics of Gibbons et al. (1989) (referred to as the
GRS statistics) and their p-values for testing the hypothesis that the alphas for the ten decile portfolios
sorted on asset growth rate are jointly zero. At the 10% critical level, the GRS test rejects the joint zeroalpha
hypothesis for five countries: Japan, Hong Kong, Malaysia, Korea, and Indonesia, as well for the allAsia
portfolios, all-Asia excluding Japan portfolios, and the U.S. market.