With three unique pairs and three coefficients, we should get a perfect fit: As I mentioned in Section 7.2.1, when there are only three categories, the quasi-variances perfectly recover the estimated variances for pairwise differences in coefficients. Demonstrate that this is the case by fitting the GLM. Some additional comments:
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here