Error in HCM model (NA in initial gradient value)
Posted: 29 May 2022, 13:41
Dear Concern,
I get the following error while running HCM+Ol with 4 random components with the Halton draws of 500. Initially, I worked with Halton draws of 300 which easily converged. In this case, the model converged but I got this error.
Any ideas to deal with such a problem?
"Additional convergence test using scaled estimation. Parameters will be scaled by their current estimates and additional
iterations will be performed.
Error in maxOptim(fn = fn, grad = grad, hess = hess, start = start, method = "BFGS", :
NA in the initial gradient"
I get the following error while running HCM+Ol with 4 random components with the Halton draws of 500. Initially, I worked with Halton draws of 300 which easily converged. In this case, the model converged but I got this error.
Any ideas to deal with such a problem?
"Additional convergence test using scaled estimation. Parameters will be scaled by their current estimates and additional
iterations will be performed.
Error in maxOptim(fn = fn, grad = grad, hess = hess, start = start, method = "BFGS", :
NA in the initial gradient"