i have been working with using maximum likelihood to estimatethe parameters of the unobserved asset process for structural risk models (no problem wit this). as part of this, i would also like to investigate the asymptotic distributions of point estimates such as asset price and credit spread which use these estimators. To do this, I attempt to follow the process outlined in (1) fordetermining the asymptotic distirbutions, i need the derivatives ofthe functions with respect to the estimators.Given a log-normal asset proess, S(t) = S(0)*exp((mu - sig^2/2)*t + sig*W(t))I need to determine d[S(t)]/d[sig], but I am stuck here. I don'tknow how to differentiate this.This is probably a basic question in a not-so basic application,but stuck is stuck.thanks in advance,--dan(1) Duan, J.C., 1994, Maximum Likelihood Estimation Using Price Data of the Derivative Contract, Mathematical Finance 4, 155-167.

I think I figured it out - and just throw it out for whatit is worth.first, please excuse my notation - in original post itwould have been more clear if i had used V - for thefirm asset process, instead of S as the basic mertonstructural credit risk model calibrates firm value basedon observered equity prices.For my question, I need d[V(sigma)]/d[sigma]. Using MLE,we have to invert the equity relationship to determineV on a point-wise basis. (-1) V = E (V, D)the sigma we are differentiating with is the estimator,so d[E]/d[sigma] = 0. Using this, and differentiatingthe equity relationship, and solving for d[V(sigma)]/d[sigma] is making more sense and seems to put me back on the right path for derivingthe work in the referenced paper by Duan.Sorry to bother the list.