Do tree based methods like random forest and gradient boosting produce unbiased estimates?

Question

Could anyone point me to literature that discuss properties of tree based estimators? For example, are they unbiased, consistent, maximum likelihood, efficient, etc?

Related: [Does LASSO suffer from the same problems stepwise regression does?](https://stats.stackexchange.com/questions/411035/does-lasso-suffer-from-the-same-problems-stepwise-regression-does) — Alexis, Mar 04 '20 at 22:08

score 1 · Accepted Answer · answered May 04 '18 at 16:27

For random forests, Hemant Ishwaran (https://scholar.google.com/citations?hl=en&user=5PSPBOEAAAAJ) has published a few interesting papers with theoretical treatments of random forests, take this for example, on the consistency of Random Survival Forests: https://www.sciencedirect.com/science/article/pii/S0167715210000672

I would also recommend looking into these:

Lucas Mentch, Giles Hooker (2016) Quantifying Uncertainty in Random Forests via Confidence Intervals and Hypothesis Tests; Journal of Machine Learning Research 17(26):1−41.
G. Biau, L. Devroye, G. Lugosi (2008) Consistency of random forests and other classifiers Journal of Machine Learning Research, 9, pp. 2039-2057

thanks for sharing these articles. still looking for some info on unbiasedness and gradient boosting algorithms, if possible. — Nandi Subhrangshu, May 07 '18 at 19:52

Do tree based methods like random forest and gradient boosting produce unbiased estimates?

1 Answers1