Switched from patsy for formulaic for linear constraint translation when using the formula interface. This allows formulas to be specified using a dictionary of constraints in addition to the standard single string of list of strings (GH439, GH440).
Fixed a bug when using escaped variable names, e.g.,
y ~ 1 + `x 3`, in panel data models (GH435).
Fixed a bug that affected creating IV models using formulas with missing data (GH438).
Fixed examples in documentation (GH434).
Clarified the null in the F-statistic
Improved the error message when dependent and exog have different numbers of observations
Added formulaic as the preferred formula parser
Fixed a bug in
ACCovarianceestimator where the number of observations was incorrectly overwritten
Fixed a bug in
linearmodels.panel.results.PanelResults.corr_squared_within()where the correlation was not squared.
Fixed a bug that affected
SURwhen estimating models using
method="gls"and heteroskedasticity-, hac, or cluster-robust inference.
rank_checkargument to panel-data models that allows the rank check to be skipped. Estimating a model that is rank deficient may result in unreliable estimates and so caution is needed if using this option.
Changed the default least squares used to
scipy.linalg.lstsq()so that the
lapack_drivercan be changed to use QR factorization.
Correct calculation of first-stage F-statistic in IV models.
Minor release to fix a wheel-building issue on Python 3.9
Improved performance of
fit()by deferring some operations.
Added support for the method available in PyHDFE in
AbsorbingLS. These methods can only be used when the variables absorbed are categorical (i.e., fixed-effects only) and when the model is unweighted.
Added a clustered covariance estimator (
linearmodels.system.covariance.ClusteredCovariance) for system regressions (GH241).
AbsorbingLSwhich allows a large number of variables to be absorbed. This model can handle very high-dimensional dummy variables and has been tested using up to 1,000,000 categories in a data set with 5,000,000 observations.
Fixed a bug when estimating weighted panel models that have repeated observations (i.e., more than one observation per entity and time id).
Added optional Cythonized node selection for dropping singletons
Added preconditioning to the dummy variable matrix when
fit(). In models with many effects, this can reduce run time by a factor of 4 or more.
Added an option to drop singleton observations in
PanelOLSby setting the keyword argument
False, singelton observations are dropped before the model is fit, so the the result is as-if the observations were never in
Added a method to construct the 2-core graph for 2-way effects models, which allows singleton observations with no effect on estimated slopes to be excluded. (GH191)
Fixed a bug where IV models estimated with only exogenous regressors where not being correctly labeled as OLS models in output. (GH185)
wald_testto panel-model results.
Added a low-memory option to
fit()that avoids constructing dummy variables. Only used when both
True. By default, the low memory algorithm will be used whenever constructing the dummy variable array would require more than 1 GiB. (GH182)
Fixed a bug which failed to correctly check rank conditions when specifying asset pricing models (GH173)
Switched to external package cached-property to manage caching instead of custom and less-well-tested solution (GH172)
Fixed a bug where weights were incorrectly calculated for HAC covariances when the weight function was
Changed the return type of Wooldridge’s over identification test when invalid to
Add typing information to IV models
Allow optimization parameters to be passed to
Removed internal use of pandas Panel
Improved performance in panel models when using
Switched to retaining index column names when original input index is named
Modified tests that were not well conceived
Added spell check to documentation build
Improve docstring for
Corrected bug that prevented single character names in IV formulas
Corrected kappa estimation in LIML when there are no exogenous regressors
Improved performance of Panel estimators by optimizing data structure construction
Added a license
Added System GMM estimator
Added automatic bandwidth for kernel-based GMM weighting estimators
Cleaned up HAC estimation across models
predictmethod to IV, Panel and System model to allow out-of-sample prediction and simplify retrieval of in-sample results
Fixed small issues with Fama-MacBeth which previously ignored weights
Added Seemingly Unrelated Regression (SUR) Estimator
Added Three-stage Least Squares (3SLS) Estimator
Added Fama-MacBeth estimator for panels
Added linear factor models for asset pricing applications
Time-series estimation using traded factors
2- and 3-step estimation using OLS
Added panel models – fixed effects, random effects, between, first difference and pooled OLS.
Addition of two-way clustering to some of the IV models (2SLS, LIML)
Added Instrumental Variable estimators – 2SLS, LIML and k-class, GMM and continuously updating GMM.