search

s3alfisc / fwildclusterboot

Fast Wild Cluster Bootstrap Inference for Regression Models / OLS in R. Additionally, R port to WildBootTests.jl via the JuliaConnectoR.
https://s3alfisc.github.io/fwildclusterboot/
GNU General Public License v3.0
22 stars 4 forks source link

bayesian bootstrap #118

Closed soodoku closed 1 year ago

soodoku commented 1 year ago

Dear Alex,

There are some collinearity issues that cause fwildclusterboot to fail. I think the options include --- ignore those samples (and report on those separately) or support bayesian bootstrap: https://twitter.com/instrumenthull/status/1487469316010389516?lang=en

s3alfisc commented 1 year ago

Hi @soodoku , thanks for reporting!

Do you have an example at hand so that I can reproduce this? Is this related to fixest dropping covariates in case of multicollinearity? My assumption was actually that collinearity should be handled smoothly (i.e. if the OLS package does not complain, fwildclusterboot shouldn't either. If it does, I would assume that something is off in my data processing scripts?

Fyi, in case you are not aware - I think the bayesian bootstrap is nowadays implemented in sandwich(link) and via the fwb package.

soodoku commented 1 year ago

@s3alfisc I miscommunicated the problem. Scratch this. Thank you getting back!