New GitHub Release
Live on GitHub!
The gfoRmulaICE package implements iterative conditional expectation (ICE) estimators of the plug-in g-formula (links included below).
Both singly robust and doubly robust ICE estimators based on parametric models are available.
The package can be used to estimate survival curves under sustained treatment strategies (interventions) using longitudinal data with time-varying treatments, time-varying confounders, censoring and competing events. The interventions can be static or dynamic, and deterministic or stochastic (including threshold interventions). Both prespecified and user-defined interventions are available.
Earlier this year, CAUSALab also published the pygformula package. This comprehensive package is the first to implement the g-formula in Python. Development team led by Postdoctoral Fellow Jing Li.
The pygformula package implements the non-iterative conditional expectation (NICE) estimator of the g-formula algorithm. The g-formula can estimate an outcome’s counterfactual mean or risk under hypothetical treatment strategies (interventions) when there is sufficient information on time-varying treatments and confounders.
Access the releases on CAUSALab’s GitHub page.
