0

I am wondering if random forest models require an independent observation assumption.

My date includes observations from the same participants, but I do not have a way to identify each participant. I would like to analyze this using random forests or decision tree, but I am not sure if I have to worry about the independent assumption like regression models or not.

I would appreciate your guidance and literature would be of great help.

user8460166
  • 596
  • 4
  • 15
  • I do not think that having independent observations is essential. The main theoretical results on random forests rely on the law of large numbers and that works for correlated data. – stans Nov 06 '19 at 06:38
  • Thank very much for your comment, @stans-ReinstateMonica! I am happy to hear that it seems like I can use random forests with my data! Do you happen to know any articles that I can cite? I would like to back up my analysis choice for writing a journal article. :) – user8460166 Nov 06 '19 at 17:55
  • Sorry, do not know of articles. I suggest examining standard references and the articles mentioned therein. – stans Nov 06 '19 at 18:01
  • Thank you for your reply! Okay, I will further read into it. I hope you have a great day! – user8460166 Nov 06 '19 at 18:07
  • Thank you. You too. – stans Nov 06 '19 at 18:32

0 Answers0