Hi, I am trying to test whether the maximum likelihood (ML) estimates of a parameter from the the results of running 'optim', are different. After searching the R-help archives, I have seen only 2 prior posts where a less detailed question has been asked about "general hypothesis testing after 'optim'", and a response offered that a likelihood ratio test (LRT) could work--However an LRT would be more applicable for comparing nested models I think. In my situation I have two point estimates and I would like to know if they are significantly different from each other. From 'optim', you can pull out the hessian and extract the variance about a given estimate, but thats as far as I have gotten. Are there any resources/ideas on how to perform an appropriate test on these ML parameter estimates? Any help will be much appreciated! Cheers, LP [[alternative HTML version deleted]]