Derive the maximum-likelihood estimating equations for the binomial logit model. Show that this model produces the same estimated coefficients as the dichotomous (binary) logit model of Section 14.1. (Hint: Compare the log-likelihood for the binomial model with the log-likelihood for the binary model; by separating individual observations sharing a common set of X-values, show that the former log-likelihood is equal to the latter, except for a constant factor. This constant is irrelevant because it does not influence the maximum-likelihood estimator; moreover, the constant disappears in likelihood-ratio tests.)
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here