Ivmodels

------------------

**Bug fixes:**

- The :class:`~ivmodels.summary.Summary` now correctly includes the rank and J test results.

------------------

**Bug fixes:**

- Fixed the ``setuptools`` configuration.

------------------

**New features:**

- The Wald test now supports robust covariance estimation.

- New method ``length`` for :class:`~ivmodels.confidence_set.ConfidenceSet`.

**Other changes:**

- One can now pass the tolerance parameter ``tol`` to the optimization algorithm in
:func:`~ivmodels.tests.lagrange_multiplier.lagrange_multiplier_test` and
:func:`~ivmodels.tests.lagrange_multiplier.inverse_lagrange_multiplier_test` via the
``kwargs``.

- :class:`~ivmodels.models.kclass.KClass` now raises if ``kappa >= 1`` (as for the
LIML and TSLS estimators) and the number of instruments is less than the number of
endogenous regressors.

- The :class:`~ivmodels.summary.Summary` now only includes and prints the results of the
J-statistic and (multivariate) F-test for instrument strength if this makes sense.

- The docs have been updated and include examples.

------------------

**New features:**

- New test :func:`~ivmodels.tests.j.j_test` of the overidentifying restrictions.

- The tests :func:`~ivmodels.tests.lagrange_multiplier.inverse_lagrange_multiplier_test`
and
:func:`~ivmodels.tests.conditional_likelihood_ratio.inverse_conditional_likelihood_ratio_test`
now possibly return unions of intervals, instead of one conservative large interval.

**Bug fixes:**

- Fixed bug in :func:`~ivmodels.models.kclass.KClass.fit` when ``C`` is not ``None`` and
:math:`M_{[Z, C]} X` is not full rank.

- Fixed bug in
:func:`~ivmodels.tests.conditional_likelihood_ratio.inverse_conditional_likelihood_ratio_test`
when ``k == mw + mx`` and ``C`` is not ``None``.

- Fixed bug in :func:`~ivmodels.utils._characteristic_roots` if
``b == np.array([[0]])``. This now correctly returns ``np.inf``.

**Other changes:**

- The :class:`~ivmodels.summary.Summary` now additionally reports the LIML variant of
the J-statistic.

------------------

**Bug fixes:**

- Fixed bug in
:class:`~ivmodels.tests.conditional_likelihood_ratio.inverse_conditional_likelihood_ratio_test`.

------------------

**New features:**

- New functions :func:`~ivmodels.tests.inverse_lagrange_multiplier_test` and
:func:`~ivmodels.tests.inverse_conditional_likelihood_ratio_test` to approximate the
1 dimensional confidence sets by inverting the corresponding (subvector) tests.

- New class :class:`~ivmodels.confidence_sets.ConfidenceSet`.

- New class :class:`~ivmodels.summary.Summary` holding information about the model fit.

- New class :class:`~ivmodels.summary.CoefficientTable` holding a table of coefficients
and their p-values.

- New method :func:`~ivmodels.models.kclass.KClass.summary` to create a summary of the
model fit.

- The :class:`~ivmodels.models.kclass.KClass` gets new attributes after fitting a model:
``endogenous_names_``, ``exogenous_names_``, and ``instrument_names_``. If pandas is
installed, there's also ``names_coefs_``.

- The tests :func:`~ivmodels.tests.anderson_rubin_test`,
:func:`~ivmodels.tests.lagrange_multiplier_test`,
:func:`~ivmodels.tests.likelihood_ratio_test`, and
:func:`~ivmodels.tests.wald_test` and their inverses
:func:`~ivmodels.tests.inverse_anderson_rubin_test`,
:func:`~ivmodels.tests.inverse_lagrange_multiplier_test`,
:func:`~ivmodels.tests.inverse_likelihood_ratio_test`, and
:func:`~ivmodels.tests.inverse_wald_test` now support an additional parameter ``D``
of exogenous covariates to be included in the test. This is not supported for
the conditional likelihood ratio test.

**Other changes:**

- The function :func:`~ivmodels.tests.lagrange_multiplier_test` is now slightly faster.

- :class:`~ivmodels.models.kclass.KClass` now accepts ``pandas.Series`` as arguments to
``y``.