Is there any package in R that we can conduct regression analysis while for instance we also correct for heteroscedasticity. I know you can use lm() function and consequently sandwich(). However, I am looking for something that I can do all the works in one place.
Hi, and welcome!
Take a look at this Czar Yobero post from a few years ago about using the generalized least squares estimator to correct a model to achieve a constant error variance and running
lm with robust standard errors.
Look at the packages descriptions in the Robust Task View.
And search rseek for
If you're interested in directly modeling the heteroscedasticity, check out the the arxiv link to Paul's paper.
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.