S3E17: Logistic Regression: 2 Logit 2 Quit
Quantitude
00:00
Using a Linear Model to Predict Weight
In the linear model, we have a sum of squares associated with an outcome. That is essentially how bad your doing at prediction if you had just predicted a mean weight for everybody. But that gets a lot trickier when we move into the maximum likelihood world. So tell me how do we use the magic of maximum likelihood to get kind of r squared by kinda, you mean, a fake or that's what we should call it?
Transcript
Play full episode
