#statstab #433 The Table 2 Fallacy: Presenting and Interpreting Confounder and Modifier Coefficients

Thoughts: Adding predictors to your model results in marked changes in inference. Be mindful!

#table2 #fallacy #covariates #bias #inference #regression

https://doi.org/10.1093/aje/kws412

'Wasserstein F-tests for Frechet regression on Bures-Wasserstein manifolds', by Haoshu Xu, Hongzhe Li.

http://jmlr.org/papers/v26/24-0493.html

#wasserstein #covariates #covariate

'Decorrelated Variable Importance', by Isabella Verdinelli, Larry Wasserman.

http://jmlr.org/papers/v25/22-0801.html

#regression #covariates #prediction

Decorrelated Variable Importance

'Scalable high-dimensional Bayesian varying coefficient models with unknown within-subject covariance', by Ray Bai, Mary R. Boland, Yong Chen.

http://jmlr.org/papers/v24/20-1437.html

#lasso #mcmc #covariates

Scalable high-dimensional Bayesian varying coefficient models with unknown within-subject covariance

Later on in drug development, as we get close to registration, we can use these models to identify #Covariates whcih might inform differences in exposure and effect between patients (like age, weight, and sex), and to quantify the relationships between dose, exposure, and response for efficacy (e.g. how well the drug does at reducing or eliminating a tumour) and safety (e.g. how many unwanted side effects the drug generates at a useful dose).