Generalized Linear Models I

The basics: how GLMs differ from linear models, what link functions are about, and how to choose among them

[@johnpvanek]

I've taken many stats courses and this is the most helpful by far!

[@christinadereklawrence1442]

I'm a bit confused when to do a GLM with a log link vs just a log transform of Y. 11:29 suggests that if data have > variance with > mean, use log-transform of Y. However, the chart at 19:53 says that log link is appropriate when variance ~ mean, which is just another way of saying that variance gets bigger than mean, which contradicts the first point. What am I missing?

[@yaxinliu5951]

Thank you so much! I finally understand how to distinguish each distributions and their properties and functions! This is very helpful, thanks for your work!!

[@smallypuppy22]

I just wanna make emphasis on this, this is the first time I have been able to understand this as well

[@KnorpelDelux]

Vielen Dank, thanks so much from overseas. :)

[@MegaSarju]

Very nice. Thanks for sharing...

[@wlancer8826]

Literally save my life!! Many thanks!

[@fabiovargasbr]

Good stuff. Well explained.

[@niceday2015]

Thank you very much

[@FRANCESCO-wj8rs]

Hello, what are the advantages of GAMLSS over GLMs?

[@HxCWW]

Why do we take ln(y) in the log link and not the ln(mean(y)) at 8:31?

[@danielawilner8173]

Thank you so much! This is so so so helpful!

[@Jojo-hz6rk]

This was just awesome...