Gauss-Markov-Style meme
from merari42@lemmy.world to science_memes@mander.xyz on 27 Sep 22:07
https://lemmy.world/post/36565423

#science_memes

threaded - newest

Bishma@discuss.tchncs.de on 27 Sep 23:28 next collapse

Yo, listen up, here’s a story
About a little guy that lives in a BLUE world
And all day and all night and everything he sees is just Best Linear Unbiased Estimator

Da-ba-dee, da-ba-di, da-ba-dee, da-ba-di

benni@lemmy.world on 28 Sep 03:06 next collapse

I don’t get it, can you explain?

montechristo@feddit.org on 28 Sep 09:45 collapse

To the practicing statisticians out there: Is BLUE at all relevant in the field? Even if you can formulate your data/problem as a linear regression, wouldn’t maximum likelihood estimation still be the preferred method in most scenarios? Edit: Of course Bayesian analysis is of equal relevance here!

merari42@lemmy.world on 28 Sep 14:05 collapse

OLS is well used in a lot of domains. If you have a continuous outcome you have a straightforward and very easy way to get a quick reliable estimate of some interesting parameter. The typical use in most social sciences is to implement some sort of difference-in-differences (or DiD event-study) estimate within a simple linear regression.