Lecture 20

• After discussing R^2 in the linear regression model (explained variance/total variance), we started looking at the maximum likelihood estimator (MLE; 6.10). There are two steps in finding the MLE;

1. find the likelihood function
2. find the parameter that maximizes the likelihood function.

For convenience, we often use the log-likelihood function to take derivatives.

• Homework 5 answer key will be posted after Friday.