linearmodels.iv.results.IVModelComparison¶

class linearmodels.iv.results.IVModelComparison(results: Sequence[IVResults | IVGMMResults | OLSResults] | dict[str, IVResults | IVGMMResults | OLSResults], *, precision: str = 'tstats', stars: bool = False)[source]¶

Comparison of multiple models.

Parameters:¶

results: Sequence[IVResults | IVGMMResults | OLSResults] | dict[str, IVResults | IVGMMResults | OLSResults]¶

Set of results to compare, which are IVResults, IVGMMResults or OLSResults. If a dict, the keys will be used as model names.

precision: str = 'tstats'¶

Estimator precision estimator to include in the comparison output. The default is “tstats”. The names are not case sensitive.

stars: bool = False¶

Add stars based on the p-value of the coefficient where 1, 2 and 3-stars correspond to p-values of 10%, 5% and 1%, respectively. The default is False.

Raises:¶

• TypeError – If a result is not an IVResults, IVGMMResults or OLSResults.

• ValueError – If precision is not supported.

Methods

Properties

cov_estimator	Covariance estimator descriptions.
estimator_method	Estimation methods.
f_statistic	F-statistics and P-values of all models.
nobs	Number of observations used to estimate each model.
params	Parameter estimates of all models.
pvalues	Parameter p-values of all models.
rsquared	Coefficients of determination (R**2).
rsquared_adj	Sample-size adjusted coefficients of determination (R**2).
std_errors	Parameter standard errors of all models.
summary	Model estimation summary.
tstats	Parameter t-statistics of all models.