https://t.co/R8pPYTfU4F I’m not convinced. In causal modelling, we want to include covariates only if they are confounders in the causal relationship of interest. There is no statistical test for confounding, so this algorithm is invalid for the purposes o
2,402 followers
338 followers
“Any variable having a significant univariate test at some arbitrary level is selected as a candidate for the multivariate analysis. We base this on the Wald test from logistic regression and p-value cut-off point of 0.25.” https://t.co/4ovuFbSIPg
3,902 followers
@dutchiewhocodes Congratulations! I love modeling in Tidyverse. Here is my favorite paper about model-fitting that guides you on what independent variables to keep: https://t.co/DGU52YOCSq
240 followers
RT @AleksiReito: What is "purposeful" selection of variables in multivariable models? From this article: https://t.co/hGia84MPb7👇 "The pur…