How it works...
There are essentially three types of robust estimates for regression models: M-estimators, S-estimators, and MM-estimators. MM and S estimators are variations of the M estimators that we presented in the first recipe. A thorough discussion about the three of them can be found here: https://ijpam.eu/contents/2014-91-3/7/7.pdf.
Anova.lmrob is just a wrapper that conveniently uses the output produced by lm.rob models. The lm.rob function uses an MM-estimator, which is highly robust (it has a 50% breakdown point – meaning that we can contaminate 50% of the data and still get reasonable results), and it is highly efficient (the variance for the estimated coefficients is as large as the one obtained via linear regression).
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access