Talk:Freedman's paradox

Why can it happen?

The article needs to explain why “predictor variables with no explanatory power can appear artificially important”. And, is this phenomenom specific to certain model selection techniques? If so, is the estimator biased, is the coefficient’s standard error biased, or what? Loraof (talk) 21:49, 15 May 2018 (UTC)

What's a regressor?

This article is not only a stub, it's also full of meaningless technical speech. The minimum that needs to be done is to provide wikilinks for technical terms like 'regressor', although actually it's possible for the second paragraph in an article to be provided using simplified, less-technical (or non-technical) language much like the lead. This stub/article reads like it comes from a textbook where the author assumes that the student will be receiving lessons from a teacher who can clarify the gibberish and incomplete information contained in the text. A stub page is not really a first draft of an article. It's valid for a stub page to contain _only_ a lead and a notice that it's a stub. Incomplete tech speech does not really clarify anything for the reader and can just as well be left out of a stub page. The author of this page needs to get it into a publishable condition or remove the mumbo jumbo. Wikipedia is an encyclopedia, not an academic journal. Comiscuous (talk) 05:20, 23 March 2021 (UTC)