New g-computation estimators, expressed as stacked estimating equations, recover average causal effects under treatment-induced selection and under confounding plus selection bias when no single adjustment set exists.
Delicatessen: M-Estimation in Python
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
M-estimation is a general statistical framework that simplifies estimation. Here, we introduce delicatessen, a Python library that automates the tedious calculations of M-estimation, and supports both built-in user-specified estimating equations. To highlight the utility of delicatessen for quantitative data analysis, we provide several illustrations common to life science research: linear regression robust to outliers, estimation of a dose-response curve, and standardization of results.
citation-role summary
background 1
citation-polarity summary
fields
stat.ME 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Constructing g-computation estimators: two case studies in selection bias
New g-computation estimators, expressed as stacked estimating equations, recover average causal effects under treatment-induced selection and under confounding plus selection bias when no single adjustment set exists.