Laurent Bergé

Original data set cannot be safely accessed when the estimation is performed in a loop

7

### Example ```R library(fixest) base = setNames(iris, c("y", "x1", "x2", "x3", "species")) all_species = unique(base$species) res_all = list() for(i in 1:3){ s = all_species[i] res_all[[i]] = feols(y ~ x1 +...

First thanks for the interest in the package and for wanting to submit a bug report. This feedback is invaluable since it significantly improves software quality, benefiting to all users....

🔥 Major bug introduced in fixest 0.12.0 for R versions 4.1.0 to 4.1.2 -- SOLVED

2

`fixest` 0.12.0 introduced a major *internal* retooling whereby most print and string manipulation tools were offloaded to the package `stringmagic`. after upgrading to `fixest` 0.12.0 several users reported the following...

From a discussion in #491 > In Clubsandwich they use a Satterthwaite approximation in the denominator of the F statistic which gives far more reliable estimates. https://jepusto.github.io/clubSandwich/articles/Wald-tests-in-clubSandwich.html