Interpreting Significance in the Fama–MacBeth Procedure

Summary

The document asks how the Fama–MacBeth procedure accounts for uncertainty in its final average coefficient. The procedure estimates regressions across periods and averages the resulting coefficients; the question is whether inference based on variation among those estimates properly reflects the fact that each coefficient is itself estimated with error.

An illustrative case imagines several individually imprecise coefficients with similar positive values. Treating those estimates as fixed would make their average seem precise, prompting concern that the reported significance could be misleading. The document does not provide an answer or evidence, so it is best read as a statistical question about the distinction between variation across period estimates and estimation uncertainty within each regression. It gives no specification details, sample structure, or inference results from which to resolve the issue; the appropriate interpretation depends on the assumptions and variance estimator used in an actual application.

Key ideas

• Fama–MacBeth estimates an overall coefficient by averaging regression coefficients across periods.
• The document questions how inference accounts for uncertainty in each period-specific estimate.
• Similar coefficient estimates may produce a precise-looking average even when individual estimates are imprecise.
• The discussion poses a methodological question but does not supply a resolution or empirical evidence.

Tags

# Fama McBeth - Significance


# Fama McBeth - Significance












The very last step of the Fama McBeth procedure is to aggregate the estimated regression coefficients by taking their mean. The mean is then the estimate for the "overall" regression coefficient.

The significance of this mean is derived by looking at the variation of the regression coefficients. But by doing so, don't we ignore the fact that the different regression coefficients are not fix?

If I take the mean of three regression coefficients that are not significantly different than zero, but which's values are let's say 10, 11 12. It is obvious that treating them as fixed at 10, 11 and 12 would yield to the conclusion that the mean (11) is different from zero with a high significance. However, this stands in contrast to the fact that each of the single coefficients is not significantly different from zero.

Is my thinking correct? If not - how is the fact that the estimates are not fix considered in the Fama McBeth procedure

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.