Arthur Chatton

Arthur Chatton

Postdoctoral researcher in Biostatistics

Université de Montréal, Canada

Bonjour !

I’m a French postdoctoral researcher interested in both prediction and causation. My research stands at the crossroads of these two worlds.

A lot of causal estimators have been developed but little guidance exists about when use (or not) them. The heart of my PhD was to compare the performances of such estimators in various context through simulations, and how mixing them with machine learning. But causal inference is not only a statistical/estimation problem. Identifiability of causal effects is also a crucial challenge in modern research. Thereby, my current research aims to develop a tool to check positivity in various context (e.g., mediation or longitudinal settings) non-parametrically. Machine learning is also a powerful tool for prediction. I’m currently developing a longitudinal predictive score for optimizing dialysis’ sessions thanks to an IVADO postdoctoral fellowship.

Interests: Causal inference, Methodology, Prediction, R Programming, Simulation study

Recent publications

Causal inference in case of near-violation of positivity: comparison of methods
Causal and Associational Language in Observational Health Research: A Systematic Evaluation
We estimated the degree to which language used in the high-profile medical/public health/epidemiology literature implied causality …
Donor HLA Class 1 Evolutionary Divergence Is a Major Predictor of Liver Allograft Rejection
SMILE: a predictive model for Scoring the severity of relapses in MultIple scLErosis
In relapsing–remitting multiple sclerosis (RRMS), relapse severity and residual disability are difficult to predict. Nevertheless, this …