Hello,
It is possible to use 'deriv' when the expression itself is dynamic? I have
data where the conditional mean is time-varying (like a GARCH(1,1) model) as
mu_{t} = omega1 + alpha1*N_{t-1} + beta1*mu_{t-1}.
The parameter vector is c("omega1", "alpha1", "beta1") and N_t is the
observation at each time.
I want two things:
1. the derivative of mu_t with respect to each parameter. So for example, the
first derivative with respect to omega1 is "1 + beta1* d mu_{t-1}/ d omega1."
2. Once I have the expression in 1, I want its first partial derivative.
Ultimately, I want to find the root of an estimating function using
Newton-Raphson, where
param <- param - solve(mat)%*% param,
and where param is an estimating function which has first partial derivatives
in it and 'mat' is a matrix which involves the first partial derivatives of
elements in 'param'.
Thanks,
Melody
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
