missing-values

Generalizing treatment effects with incomplete covariates

We focus on the problem of generalizing a causal effect estimated on a randomized controlled trial (RCT) to a target population described by a set of covariates from observational data. Available methods such as inverse propensity weighting are not …

Causal Survival Analysis from Theory to Practice

Causal survival analysis consists in estimating the effect of a treatment on time-to-event outcome(s). We focus on estimating the restricted mean survival time (RMST), the average survival time from baseline to a pre-specified time, between treated …

Treatment effect estimation with incomplete attributes

Missing attributes are ubiquitous in causal inference, as they are in most applied statistical studies. In this work, we consider various sets of assumptions under which causal inference is possible despite missing attributes and discuss …

Causal inference methods for combining randomized trials and observational studies

With increasing data availability, treatment causal effects can be evaluated across different datasets, both randomized trials and observational studies. Randomized trials isolate the effect of the treatment from that of unwanted (confounding) …

Transporting treatment effects with incomplete attributes

The simultaneous availability of experimental and observational data to estimate a treatment effect is both an opportunity and a statistical challenge: Combining the information gathered from both data is a promising avenue to build upon the internal …

R package: misaem is back on CRAN

Estimate parameters of linear regression and logistic regression with missing covariates with missing data, perform model selection and prediction, using EM-type algorithms. Jiang W., Josse J., Lavielle M., TraumaBase Group (2020).

Inférence causale pour données observationnelles et analyses jointes de données expérimentales et observationnelles

Dans le domaine de l’apprentissage machine, de grands progrès ont été réalisés dans l’obtention de modèles prédictifs puissants, mais ces modèles reposent sur des corrélations entre variables et ne permettent pas de comprendre les mécanismes …

ICML workshop: The Art of Learning with Missing Values

We are very pleased to announce a workshop at this year's ICML conference on The Art of Learning with Missing Values, co-organized by Julie Josse and Gaël Varoquaux.

Causal inference for observational clinical data

In machine learning, there has been great progress in obtaining powerful predictive models, but these models rely on correlations between variables and do not allow for an understanding of the underlying mechanisms or how to intervene on the system …

Treatment effect estimation with missing attributes

Postponed to 2021 due to COVID-19.