import pyfixest as pf
fit = pf.feols("Y ~ X1 | f1", pf.get_data(), vcov="HC1")
table = fit.coeftable
table.estimate, table.se, table.conf_int(array([-0.94944106]),
array([0.06736535]),
array([[-1.08164036],
[-0.81724176]]))
Coefficient estimates with their standard errors, statistics, and bounds.
get_inference() publishes one value as fit.coeftable from the current fit.variance_covariance; it is the table that fixest’s coeftable() prints, plus the confidence bounds of confint().
| Name | Type | Description | Default |
|---|---|---|---|
| estimate | NDArray[np.float64] | Coefficient estimates, shape (n_coefficients,). | required |
| se | NDArray[np.float64] | Standard errors, the square root of the covariance diagonal. | required |
| tstat | NDArray[np.float64] | estimate / se. |
required |
| pvalue | NDArray[np.float64] | Two-sided p-values from the model’s reference distribution with variance_covariance.df_t degrees of freedom. |
required |
| conf_int | NDArray[np.float64] | Lower and upper confidence bounds, shape (2, n_coefficients), at level 1 - alpha. |
required |
| alpha | float | Significance level of the bounds. | required |