glmfit covariance matrix

I was wondering, when doing a glmfit using the probit link, is the negative inverse of the estimated covariance matrix (stats.covb) equal to the Hessian matrix of the log-likelihood function evaluated at the final parameter estimates?

Any intuition on how the estimated covariance matrix is calculated would be greatly appreciated. Thanks!

Answer 1

[the cyclist]

I don't know the answer to your question, but if you type "edit glmfit" and search for "stats.covb", you can see the calculation for yourself. Maybe you'll be able to answer your own question.