Fetch a model of class Feols from the Fixest class.
Parameters
Name
Type
Description
Default
i
int or str
The index of the model to fetch.
required
print_fml
bool
Whether to print the formula of the model. Default is True.
True
Returns
Name
Type
Description
A Feols object.
pvalue
estimation.FixestMulti_.FixestMulti.pvalue()
Obtain the p-values of the fitted models.
Returns
Name
Type
Description
pandas.Series
A pd.Series with coefficient names and p-values. The key indicates which models the estimated statistic derives from.
se
estimation.FixestMulti_.FixestMulti.se()
Obtain the standard errors of the fitted models.
Returns
Name
Type
Description
pandas.Series
A pd.Series with coefficient names and standard error estimates. The key indicates which models the estimated statistic derives from.
tidy
estimation.FixestMulti_.FixestMulti.tidy()
Return the results of an estimation using feols() as a tidy Pandas DataFrame.
Returns
Name
Type
Description
pandas.DataFrame or str
A tidy DataFrame with the following columns: - fml: the formula used to generate the results - Coefficient: the names of the coefficients - Estimate: the estimated coefficients - Std. Error: the standard errors of the estimated coefficients - t value: the t-values of the estimated coefficients - Pr(>|t|): the p-values of the estimated coefficients - 2.5%: the lower bound of the 95% confidence interval - 97.5%: the upper bound of the 95% confidence interval If type is set to “markdown”, the resulting DataFrame will be returned as a markdown-formatted string with three decimal places.
to_list
estimation.FixestMulti_.FixestMulti.to_list()
Return a list of all fitted models.
Returns
Name
Type
Description
A list of all fitted models of types Feols or Fepois.
tstat
estimation.FixestMulti_.FixestMulti.tstat()
Obtain the t-statistics of the fitted models.
Returns
Name
Type
Description
A pd.Series with coefficient names and estimated t-statistics.
The key indicates which models the estimated statistic derives from.
By calling vcov() on a “Fixest” object, all inference procedures applied to the “Fixest” object are replaced with the variance-covariance matrix specified via the method.
Parameters
Name
Type
Description
Default
vcov
Union[str, dict[str, str]])
A string or dictionary specifying the type of variance-covariance matrix to use for inference. - If a string, can be one of “iid”, “hetero”, “HC1”, “HC2”, “HC3”. - If a dictionary, it should have the format {“CRV1”: “clustervar”} for CRV1 inference or {“CRV3”: “clustervar”} for CRV3 inference.
required
vcov_kwargs
Optional[dict[str, any]]
Additional keyword arguments for the variance-covariance matrix.
None
Returns
Name
Type
Description
An instance of the "Fixest" class with updated inference.
Run a wild cluster bootstrap for all regressions in the Fixest object.
Parameters
Name
Type
Description
Default
B
int
The number of bootstrap iterations to run.
required
param
Union[str, None]
A string of length one, containing the test parameter of interest. Default is None.
None
cluster
Union[str, None]
The name of the cluster variable. Default is None. If None, uses the self._clustervar attribute as the cluster variable. If the self._clustervar attribute is None, a heteroskedasticity-robust wild bootstrap is run.
None
weights_type
str
The type of bootstrap weights. Either ‘rademacher’, ‘mammen’, ‘webb’, or ‘normal’. Default is ‘rademacher’.
'rademacher'
impose_null
bool
Should the null hypothesis be imposed on the bootstrap dgp, or not? Default is True.
True
bootstrap_type
str
A string of length one. Allows choosing the bootstrap type to be run. Either ‘11’, ‘31’, ‘13’, or ‘33’. Default is ‘11’.
'11'
seed
Union[str, None]
Option to provide a random seed. Default is None.
None
k_adj
bool
Whether to adjust the original coefficients with the bootstrap distribution. Default is True.
True
G_adj
bool
Whether to adjust standard errors for clustering in the bootstrap. Default is True.
True
Returns
Name
Type
Description
pandas.DataFrame
A pd.DataFrame with bootstrapped t-statistic and p-value. The index indicates which model the estimated statistic derives from.