Testing Fama–French Factors Out of Sample and Choosing Test Portfolios

# Does Fama French Three Factor Model Work out of Sample (after 1993)?


# Does Fama French Three Factor Model Work out of Sample (after 1993)?












Does anyone know if the Fama-French three factor model has been re-examined empirically after 1993, when the original paper was first published?

I am asking because there seems to be considerable debate about whether the SMB and HML factors represent true risk-factors. It seems to me the best test is to see whether they still work out of sample.

## Answer by vonjd (score 7)

https://quant.stackexchange.com/a/8597

You might want to read this:

Size, Value, and Momentum in International Stock Returns by Fama and French (2011)

> Abstract: In the four regions (North America, Europe, Japan, and Asia Pacific) we examine, there are value premiums in average stock returns that, except for Japan, decrease with size. Except for Japan, there is return momentum everywhere, and spreads in average momentum returns also decrease from smaller to bigger stocks. We test whether empirical asset pricing models capture the value and momentum patterns in international average returns and whether asset pricing seems to be integrated across the four regions. Integrated pricing across regions does not get strong support in our tests. For three regions (North America, Europe, and Japan) local models that use local explanatory returns provide passable descriptions of local average returns for portfolios formed on size and value versus growth. Even local models are less successful in tests on portfolios formed on size and momentum.

## Answer by Mayou (score 4)

https://quant.stackexchange.com/a/8704

In the following paper: "On the Cross-Section of Expected Stock Returns: Fama-French Ten Years Later" (by Chou, Chou, and Wang), the authors found, using the Fama-Mac Beth two-pass regression, that the size effect becomes insignificant during the post-1981 period, and the Book/Market effect becomes insignificant during the post-1990 period.

It is important to note that the statistical significance of the Fama-French factors is not only highly sensitive to the sample period used for testing, but also highly dependent on the test assets used as dependent variables. In both Fama-French papers (1993, 2011), the test assets were double-sorted portfolios formed on size and book/market, and/or size and momentum.

The premise of Fama-French seems reasonable: if there are priced factors that are responsible for the size and value premium, then sorting stocks into portfolios based on their size and book/market ratio are likely to result in diversified portfolios that span the factor space.

The problem is that, by grouping all of the assets with similar size or B/M together, any variation in factor loading that is independent of these two firm-characteristics is largely eliminated.

Therefore, in order to have more powerful tests, the LHS(left-hand side) portfolios should be augmented by portfolios with high correlation to the proposed factors, but with imperfect correlation with size and B/M.