⚠️ IMPORTANT NOTICE: This development version of the margot package is undergoing significant refactoring as we transition to the margotverse suite of packages. This package is currently for the author’s lab use only. The package will be split into focused, single-responsibility packages including margot.core, margot.lmtp, margot.grf, margot.viz, and others. Please expect breaking changes in upcoming releases.
MARGinal Observational Treatment-effects.1
Causal inference requires balance across the treatments to be compared. In observational studies, such balance is not guaranteed; quantifying causality therefore requires careful, multi-step workflows.
The goal of margot is to enhance understanding of causality in observational research.
The package offers functions for:
- evaluating causal assumptions
- modelling time-series data
- reporting results
- performing sensitivity analyses
margot focuses on streamlining the estimation of (Marginal) Average Treatment Effects (ATT, ATE), but it also supports workflows for Heterogeneous Treatment Effects (CATE) (estimated via grf), as well as Longitudinal Modified Treatment Policies (estimated via lmtp). It has extensive graphical and reporting functions to ease burdens for understanding.