S3E17: Logistic Regression: 2 Logit 2 Quit
Quantitude
The Pseudo R Squared
In the kicker example, where we have a binary outcome, that's like predicting every kick is point seven five. So there's a baseline model that establishes essentially how much badness of fit there is in the system if you don't take any ex information into account. We can move to what i think is singularly the most confusing part of these models. We get a regression coefficient. It is a linearization of a one unit change in x as associated with a point four unit change in the loget. Oe. Who cares if we can't interpret it? That's right. Patrick, ok. Do just pseudo r square. Sttay, on task. All right.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
