Hello,
I've got a program written in S-plus which I think is converted successfully
to R with the exception of part of the opt.param function written.
In S-plus it is:
nlminb(start=x0, obj=negllgamma.f, scale=1, lower=c(0.01,0.0001),
upper=c(10,0.9999), gamma=gamma, maxlik=maxlik,
y=ldose, s=lse, max.iter = 1000, max.fcal = 1000)$par
and so far with R I've got to:
optim(par=x0, fn=negllgamma.f, method="L-BFGS-B",
lower=c(0.01,0.0001),
upper=c(10,0.9999), gamma=gamma, maxlik=maxlik,
y=ldose, s=lse, control=list(maxit = 1000))$par
however I've failed to find an equivalent to "max.fcal". I'm
not sure whether this is important of not but the program seems to fail with
certain data sets (primarily with values close to zero) so wondered if this was
the cause?
Any help would be greatly appreciated, I've got very little programming
knowledge so a layman's description would be good.
Many thanks, Helena
***************************************************
Helena Rodnight
Institute of Geography and Earth Sciences
University of Wales, Aberystwyth
Llandinam Building
Aberystwyth
Ceredigion
SY23 3DB
Tel: +44 (0)1970 622604
[[alternative HTML version deleted]]
