xtfmb - Fama and MacBeth (1973) procedure
xtfmb depvar [indepvars] [if] [in] [weight] [, level(#) verbose lag(#)]
- You must tsset your data before using xtfmb. - by, may be used with xtfmb; see by. - aweights are allowed; see weight.
xtfmb is an implementation of the Fama and MacBeth (1973) two step procedure. The procedure is as follows: In the first step, for each single time period a cross-sectional regression is performed. Then, in the second step, the final coefficient estimates are obtained as the average of the first step coefficient estimates.
If xtfmb is called without option lag(#), then it is possible to test for the significance of coefficient combinations. This works because in this case the second step of the Fama-MacBeth procedure is implemented by aid of Zellner's SUR estimation.
When xtfmb is called with option lag(#), then heteroscedasticity and autocorrelation consistent Newey-West (1987) standard error estimates are provided. However, in this case the current implementation of xtfmb does not allow for testing the significance of coefficient combinations.
The "avg. R-squared" which is provided in the header of the xtfmb program is computed as the average value of the R-squares from the cross-sectional regressions in the first step of the Fama-MacBeth procedure. The coefficient estimates and R-squares of the first step regressions can be printed out with option verbose.
level(#); see estimation options.
verbose lists the coefficient estimates and R-squares of the cross-sectional regressions from the first step of the Fama-MacBeth procedure.
lag(#) computes heteroscedasticity and autocorrelation consistent Newey-West (1987) standard error estimates with a lag length of # periods.
. webuse grunfeld . xtfmb invest mvalue kstock, verbose . est store FMB
. xtfmb invest mvalue kstock, lag(2) . est store FMB_Newey
. reg invest mvalue kstock . est store OLS
. est table *, b se t
- Fama, E.F., and J.D. MacBeth, 1973, Risk, Return, and Equilibrium: Empirical tests, Journal of Political Economy 81, 607-636.
- Newey, W.K., and K.D. West, 1987, A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix, Econometrica 55: 703Ð708.
Daniel Hoechle, University of Basel, firstname.lastname@example.org
Manual: [R] regress
Online: tsset, regress, newey, xtreg, _robust