https://github.com/bashtage/linearmodels - Version 6.1

The version improves compatibility with changes in SciPy 1.14.

- Python
Published by bashtage over 1 year ago

https://github.com/bashtage/linearmodels - Release 6.0

• Increased minimums:
  • Python: 3.9
  • formulaic: 1.0.0
  • NumPy: 1.22.3
  • SciPy: 1.8.0
  • pandas: 1.4.0
  • statsmodels: 0.13.0
• The key feature of this release is compatibility with NumPy 2. linearmodels wheels are built using NumPy 2.0.0rc1 (or later) and can run on any version of NumPy 1.22.3 or later, including NumPy 2.0.0.
• Improved compatibility with fuutre changes in pandas 3.0.0.

[!NOTE] In order to use NumPy 2, the environment must consist of packages that have been built against NumPy 2.0.0rc1 or later.

- Python
Published by bashtage almost 2 years ago

https://github.com/bashtage/linearmodels - Release 5.4

This is a compatibility release.

• Compatibility with NumPy 2
• Compatibility with recent pandas releases

- Python
Published by bashtage about 2 years ago

https://github.com/bashtage/linearmodels - Release 5.3

• Bumped the minimum formulaic to 0.6.5.
• Released wheels for Python 3.12.

- Python
Published by bashtage over 2 years ago

https://github.com/bashtage/linearmodels - Release 5.2

• Fixed a bug that affected dropping absorbed regressors when using weights (#546).
• Improved the licensing information (#545).
• Improved compatibility with recent pandas (#548).

- Python
Published by bashtage over 2 years ago

https://github.com/bashtage/linearmodels - Release 5.1

This is a bug-fix release.

• Fixes a bug that can be encountered when collection observation data in panel models when some entities or time periods are missing.

- Python
Published by bashtage over 2 years ago

https://github.com/bashtage/linearmodels - Release 5.0

This major release contains breaking changes when using formulas and increases the requirements for running linear models.

• The variable order is preserved when creating a model using from_formula. Previously variables were sorted irrespective of the order they appeared in the formula.

• Increased minimums:

  • Python: 3.9
  • formulaic: 0.6.1
  • NumPy: 1.19.0
  • SciPy: 1.5.0
  • pandas: 1.1.0
  • statsmodels: 0.12.0

• Switched variable ordering by default. Importing linearmodels.future.ordering is a no-op, and has no effect.

• Removed dependence on property-cached in favor of :meth:functools.cached_property.

- Python
Published by bashtage over 2 years ago

https://github.com/bashtage/linearmodels - Release 4.31

This release adds support for formulaic 0.6.0. This version of formulaic makes a significant change to how variables from formulas are translated into DataFrames for use in models. The existing behavior sorts variables. The new behavior respects the order of the variables as they appear. The new behavior will become the default in linearmodels starting in release 5.

To use the new behavior now, add the import

python from linearmodels.__future__ import ordering

to the top of your file.

- Python
Published by bashtage almost 3 years ago

https://github.com/bashtage/linearmodels - Release 4.30

This compatibility release addresses unreleased changes in NumPy 1.25 and improves pandas 2.0 copy-on-write behavior.

- Python
Published by bashtage almost 3 years ago

https://github.com/bashtage/linearmodels - Release 4.29

This is a feature and compatibility release.

• Added R2 measures for FamaMacBeth
• Started testing against pandas 2.0 with copy-on-write enabled

- Python
Published by bashtage almost 3 years ago

https://github.com/bashtage/linearmodels - Release 4.28

This is a maintenance and compatibility release.

• Compatibility with formulaic 0.3.2 - 0.5.2
• Compatibility with pandas 2.0

- Python
Published by bashtage almost 3 years ago

https://github.com/bashtage/linearmodels - Release 4.27

This is a bug fix and enhancement release.

• Switched from patsy for formulaic for linear constraint translation when using the formula interface.
• Fixed a bug when using escaped variable names, e.g., y ~ 1 + `x 3`, in panel data models.
• Fixed a bug that affected creating IV models using formulas with missing data.
• Fixed examples in the documentation.

- Python
Published by bashtage over 3 years ago

https://github.com/bashtage/linearmodels - Release 4.26

This is a minor release that addresses two issues:

• Fixed a bug in linearmodels.iv.model.IV2SLS and related models where linearmodels.iv.model.IV2SLS.predict would produce NaN values when exog and endog had different indices. An IndexWarning is now shown.
• Added stacklevel to all warnings to improve the accuracy of warning location.

- Python
Published by bashtage almost 4 years ago

https://github.com/bashtage/linearmodels - Release 4.25

This release adds formulaic as the formula parser and fixes some minor bugs.

• Added formulaic as the preferred formula parser
• Clarified the null in the F-statistic
• Improved the error message when dependent and exog have different numbers of observations
• Fixed a bug in ACCovariance estimator where the number of observations was incorrectly overwritten
• Fixed a bug in PanelResults.corr_squared_between, lPanelResults.corr_squared_overall, and PanelResults.corr_squared_within where the correlation was not squared.

- Python
Published by bashtage about 4 years ago

https://github.com/bashtage/linearmodels - Release 4.24

This is a feature release. There are two new features:

• SystemResults.breusch_pagan implements the Breusch-Pagan test of 0 correlation (and so a diagonal covariance)
• SystemResults.likelihood_ratio implements a likelihood ratio test of a diagonal covariance (and so 0 correlation)

- Python
Published by bashtage almost 5 years ago

https://github.com/bashtage/linearmodels - Release 4.23

This release fixes a bug that can affect standard error estimation in SUR and related models.

- Python
Published by bashtage almost 5 years ago

https://github.com/bashtage/linearmodels - Release 4.22

This is a packaging-only release that fixes an issue with Windows wheels.

- Python
Published by bashtage almost 5 years ago

https://github.com/bashtage/linearmodels - Release 4.21

This release fixes a technical issue the prevented wheels from being built. There are no other changes.

- Python
Published by bashtage almost 5 years ago

https://github.com/bashtage/linearmodels - Release 4.20

This release fixes a single bug that affected the computation of the first-stage F-statistic.

- Python
Published by bashtage almost 5 years ago

https://github.com/bashtage/linearmodels - Release 4.19

This is a minor release that fixes a wheel-building issue. There are no other changes from 4.18.

- Python
Published by bashtage about 5 years ago

https://github.com/bashtage/linearmodels - Release 4.18

This is a performance and bug-fix release.

• Improved performance of AbsorbingLS.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 alternative measures of R-squared using the squared correlation. See PanelResults.corr_squared_overall, PanelResults.corr_squared_within, and PanelResults.corr_squared_between.
• Added a clustered covariance estimator (ClusteredCovariance) for system regressions.
• Fixed a bug in kernel_optimal_bandwidth which used incorrect values for a tuning parameter in the bandwidth estimation for the Parzen and Quadratic Spectral kernels.

- Python
Published by bashtage about 5 years ago

https://github.com/bashtage/linearmodels - Release 4.17

This is a minor release that corrects multiple incorrect type definitions.

- Python
Published by bashtage about 6 years ago

https://github.com/bashtage/linearmodels - Release 4.16

This release being two features and some bug fixes:

• Added FamaMacBethResults which exposes the set of estimated regression coefficients used to construct the model estimator.
• Added typing to all functions
• Refactored code to statically type with mypy and fixed some subtle bugs

- Python
Published by bashtage about 6 years ago

https://github.com/bashtage/linearmodels - Release 4.15.4

This release fixes a small number of bugs, adds new measure of model fit, and provides some essential maintenance:

• Blackened the code.
• Added McElroy's and Berndt's measures of system fit.
• Removed support for Python 3.5 inline with NEP-29.
• Fixed a bad test that only appeared when packaging from 4.15.3.

- Python
Published by bashtage about 6 years ago

https://github.com/bashtage/linearmodels - Release 4.15.3

This release fixes a small number of bugs, adds new measure of model fit, and provides some essential maintenance:

• Blackened the code.
• Added McElroy's and Berndt's measures of system fit.
• Removed support for Python 3.5 inline with NEP-29.
• Fixed a packing error in 4.15.2

- Python
Published by bashtage about 6 years ago

https://github.com/bashtage/linearmodels - Release 4.15.2

This release fixes a small number of bugs, adds new measure of model fit, and provides some essential maintenance:

• Blackened the code.
• Added McElroy's and Berndt's measures of system fit.
• Removed support for Python 3.5 inline with NEP-29.
• Fixed a packing error in 4.15.1

- Python
Published by bashtage about 6 years ago

https://github.com/bashtage/linearmodels - Release 4.15.1

This release fixes a small number of bugs, adds new measure of model fit, and provides some essential maintenance:

• Blackened the code.
• Added McElroy's and Berndt's measures of system fit.
• Removed support for Python 3.5 inline with NEP-29.
• Fixed a packing error in 4.15

- Python
Published by bashtage about 6 years ago

https://github.com/bashtage/linearmodels - Release 4.15

This release fixes a small number of bugs, adds new measure of model fit, and provides some essential maintenance:

• Blackened the code.
• Added McElroy's and Berndt's measures of system fit.
• Removed support for Python 3.5 inline with NEP-29.

- Python
Published by bashtage about 6 years ago

https://github.com/bashtage/linearmodels - Release 4.14

This release fixes a small number of bugs and provides some essential maintenance:

• Fixed issue where datasets were not installed with wheels.
• Switched to property-cached to inherit cached property from property.
• Removed all use pandas.Panel.

- Python
Published by bashtage over 6 years ago

https://github.com/bashtage/linearmodels - Release 4.13

This is a feature and bug release.

• Added AbsorbingLS which 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 drop_absorbed option to PanelOLS which automatically drops variables that are absorbed by fixed effects.
• Added optional Cythonized node selection for dropping singletons
• Added preconditioning to the dummy variable matrix when use_lsmr=True in fit. In models with many effects, this can reduce run time by a factor of 4 or more.

- Python
Published by bashtage over 6 years ago

https://github.com/bashtage/linearmodels - Release 4.12

Highlights of this release include:

• Dropping singleton observations in PanelOLS models
• Support for LSMR as an option to estimate parameters. LSMR can be much faster in very sparse unbalanced panels.
• Added wald_test to panel model results class.
• Added a low-memory option to limit memory usage when estimating models with 2 effects.

- Python
Published by bashtage almost 7 years ago

https://github.com/bashtage/linearmodels - Release 4.11

This is primarily a bug-fix release. Two bugs which incorrectly verified rank conditions on models were fixed. These checks affected:

• All asset pricing models
• Fama-MacBeth panel data regression

In addition, an external dependency on cached-property was introduced.

- Python
Published by bashtage over 7 years ago

https://github.com/bashtage/linearmodels - Release 4.10

This is a bug-fix release that fixes an incorrect implementation of the Parzen weight function. This release also contains small fixes for future changes in dependencies.

- Python
Published by bashtage over 7 years ago

https://github.com/bashtage/linearmodels - Release 4.9

This is a minor release that focuses on fixes for upstream changes.

• Changed the return type of Wooldridge's over identification test when invalid to InvalidTestStatistic
• Add typing information to IV models
• Allow optimization parameters to be passed to IVGMMCUE
• Removed internal use of pandas Panel
• Improved performance in panel models when using from_formula
• 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 summary properties

- Python
Published by bashtage over 7 years ago

https://github.com/bashtage/linearmodels - Release 4.8

This is a minor release with one feature and one bug fix:

• Corrected bug that prevented single character names in IV formulas
• Corrected kappa estimation in LIML when there are no exogenous regressors

- Python
Published by bashtage almost 8 years ago

https://github.com/bashtage/linearmodels - Release 4.7

This is a minor release with one significant feature:

• Substantial performance improvements in PanelOLS and possibly other panel models for large dataset sizes

- Python
Published by bashtage almost 8 years ago

https://github.com/bashtage/linearmodels - Linearmodels 4.6 release

• Add a license
• Change location of documentations

- Python
Published by bashtage almost 8 years ago

https://github.com/bashtage/linearmodels - Linearmodels 4.5 release

• Added System GMM estimator
• Added automatic bandwidth for kernel-based GMM weighting estimators
• Cleaned up HAC estimation across models
• Added predict method 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

- Python
Published by bashtage about 8 years ago

https://github.com/bashtage/linearmodels - Linearmodels 4.2 Release

New features:

• System estimation using GMM when there are endogenous variables
• General clean up of HAC estimators
• Many typos fixed

- Python
Published by bashtage about 8 years ago

https://github.com/bashtage/linearmodels - Linearmodels 4.1 Release

Minor release to address PyPi and documentation issues

- Python
Published by bashtage over 8 years ago

https://github.com/bashtage/linearmodels - Linearmodels 4.0 Release

New features:

• Three-stage Least Squares (3SLS) estimation of systems of IV equations

Other:

• Make xarray optional
• Small bug and documentation fixes

- Python
Published by bashtage over 8 years ago

https://github.com/bashtage/linearmodels - Linearmodels 3.5 Release

New features:

System Estimation * Seemingly Unrelated Regression (SUR) Estimation

Other: * Remove core dependency on pandas Panel which is scheduled to be removed

- Python
Published by bashtage over 8 years ago

https://github.com/bashtage/linearmodels - Linearmodels 3.0 Release

New features:

• Linear Asset Pricing Models
  • GMM
  • 2-step
  • Seemingly Unrelated Regression for Traded Factors
• Fama-MacBeth estimator for Panel models
• Autocorrelation (only) robust estimator for Panel models

Other:

• Doc improvements

- Python
Published by bashtage almost 9 years ago

https://github.com/bashtage/linearmodels - Panel models

• Introduction of panel models -- fixed effects, random effects, between, first differecne and pooled OLS.
• Addition of two-way clustering to some of the IV models (2SLS, LIML)

- Python
Published by bashtage almost 9 years ago

https://github.com/bashtage/linearmodels - First release

The instrumental variable estimators are tested and ready. The documentation is mostly complete although there is still work to do. The next release will feature Panel data models and more documentation.

- Python
Published by bashtage almost 9 years ago