v2.3

v2.3 Release

• Checked compatibility with NumPy 2.0
• Added estimating equation for regression calibration to measurement error corrections
• Change to pharmacokinetic model underlying structure
• Added E-max model and its ED function as estimating equations
• Replaced parameter log-logistic models with the more general ee_loglogistic estimating equation. The ee_#loglogistic models will be removed in v3.0

v2.2

v2.2 release

• Added extended Rogan-Gladen to correct for differential measurement error
• Added separate functionality to compute the sandwich matrix. This avoids needing to call MEstimator to compute the sandwich matrix.
• Updated all docs

v2.1

v2.1 addition of new estimating equations: Rogan-Gladen measurement error correction, multinomial logistic regression, efficient g-estimation, log-linear SMM g-estimation.

Added support for Python 3.12

Added option to rescale spline terms when generated

Re-organized test structure for easier maintenance (does not impact actual package)

Bug fixes: fixed issue in call to ee_lasso_regression

v2.0

v2.0 adds an automatic differentiation functionality to compute the bread matrix. So, now both numerical approximation and automatic differentiation are supported.

v1.4

Additions of v1.4 release

• Added Generalized Linear Models (GLM) as a built-in estimating equation
• Added Z-scores, P-values, and S-values
• Added Marginal Structural Models with Inverse Probability Weighting as a built-in estimating equation
• Added support for missingness weights with ee_ipw and ee_gestimation_snmm

v1.3

v1.3 release

• Adds G-estimation of linear structural nested mean models
• Fixes bug with Powell hybrid method for root-finding specification

v1.2

version 1.2 release. Adds the optional argument offset to regression models in regression.py

v1.1

Release of v1.1

Adds a functionality for GAMs.

v1.0

v1.0 release

This major release changes the version support for several items. In order to speed up the computation of the bread matrix, SciPy approx_fprime funtionality is now used. This required the following version changes:

• SciPy v1.9.0+ which requires NumPy 1.18.5. This is a major change in the supported versions
• SciPy v1.9.0 also necessitates the reduction of support to 3.8+

The regression estimating equations have also now been fully updated to the new syntax. The legacy versions have been removed.

v0.6

Additional robust loss functions, better errors when nan's appear in the estimating equations, added warnings for non-differentiable estimating equations