News for hackers. Stuff that matters.

A comment on Interpretable Machine Learning. Parent comment

Re:Interpretable Machine Learning by civilized on Sunday November 29, 2020 @12:42PM (#25246610)
> This is incorrect. Linear regression does not require the outcomes follow a normal distribution. Rather, it requires that the residual errors are normally distributed.

No, the author's formulation and yours are mathematically equivalent statements, and specifying the conditional distribution of the outcomes is the version that generalizes to GLMs.

[ Parent | on HN ]
Re:Interpretable Machine Learning by fractionalhare on Sunday November 29, 2020 @12:55PM (#25246691)
Ah, I didn't see what "given" was doing in that sentence. It looks like you (and therefore the author) are correct. I concede that point :)
[ Parent | on HN ]