[R] Weighted Likelihood Estimation of NIG Dist.

n4538 j.scally at student.qut.edu.au
Wed Jan 12 10:37:35 CET 2011


Hi All,

I've put together some script which gives me the parameters of a modified
NIG distribution. Now I'd like to include a weighted vector within the
maximization function. The code below gives me parameter estimates which are
implicitly equally weighted.


     nig.par.fit <- try(optim(vega, negllh, hessian = se, pdf = density.nig,
                       tmp.data = data, transf = transform, const.pars =
vars[!opt.pars],
                       silent = silent, par.names = names(vars), ...))

vega are the parameters to be fitted.

This seems to work reasonably for my equally weighted observations, but I
have not been able to find/adjust the function to include a weighted vector
(w_i). I'm trying not to alter the density function.

Is there a way of doing this with optim, or another optimizer I should be
looking at?

Many thanks,

J
-- 
View this message in context: http://r.789695.n4.nabble.com/Weighted-Likelihood-Estimation-of-NIG-Dist-tp3213608p3213608.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list