str = 'unadjusted', debiased: bool = True, bandwidth: float | None = None, kernel: str | None = None) FamaMacBethResults[source]

Estimate model parameters

cov_type: str = 'unadjusted'

Name of covariance estimator (see notes). Default is “unadjusted”.

debiased: bool = True

Flag indicating whether to debiased the covariance estimator using a degree of freedom adjustment.

bandwidth: float | None = None

The bandwidth to use when cov_type is “kernel”. If None, it is automatically computed.

kernel: str | None = None

The kernel to use. None chooses the default kernel.


Estimation results

Return type:



>>> from linearmodels import FamaMacBeth
>>> mod = FamaMacBeth(y, x)
>>> res ="kernel", kernel="Parzen")


Two covariance estimators are supported:

  • “unadjusted”, “homoskedastic”, “robust”, “heteroskedastic” use the standard covariance estimator of the T parameter estimates.

  • “kernel” is a HAC estimator. Configurations options are: