News for hackers. Stuff that matters.
Re:Linear Regression by autokad on Friday June 19, 2020 @01:28PM (#23576771)
i wouldn't dare to say its a 'serious drawback'. as soon as you add ridge to regression (something very common for statisticians to do), it's also no longer scale/unit-invariant.

Edit: this is also not just true for ridge, but lasso as well

Re:Linear Regression by em500 on Friday June 19, 2020 @02:07PM (#23577220)
Personally I consider the lack of scale-invariance one of the main drawbacks of most common regularizers too. Again, not a big deal if all you're after are y-hat, a bit more concerning if you're interested in beta-hat.
Re:Linear Regression by autokad on Friday June 19, 2020 @05:48PM (#23579189)
i try not to stick too religiously to coefficient interpretation unless its a very simple and known problem. one missing variable could change the coefficients sign.
Re:Linear Regression by zwaps on Friday June 19, 2020 @08:06PM (#23580233)
Depends on whether the missing variable is correlated with the beta hats you care about ;-)