search for: neadler

Hello, I have been using R to fit my data using non-linear least squares. I have used the optimize routine to minimize the sum of squared errors (using Nealder Meade optimization routine), but couldn't get the non-linear model in R to converge to the estimates acheived in a convergent Neadler Meade routine. It tells me about problems with the gradient. I was wondering if there is any way to estimate the covariance matrix for the parameters when using the Nealder Meade optimization method. If so is there a book, or article, that I can refer to about how this is done, it would be great...