[R] na.include and nlm warnings

[Jukka T Jokinen]

Dear R-users,

1) has there been plans to add na.include-function for handling missing
values in objects? This would be particularly helpful for functions that
fit models for longitudinal datasets with possible missing responses.

2) I am using nlm for a numerical optimisation of a function. Sometimes
the function produces NA as a result. nlm gives a warning that NA was
replaced by a maximum positive value so, in case of NA, I replaced the
function output with .Machine$double.xmax to avoid these warnings. Since I
don't know what nlm exactly does in these situations, I am a bit worried
whether there is any chance that this might affet the numerical
optimization in some situations. Can anyone tell whether this could cause
problems?

This could be avoided with optim which doesn't print warnings in this case
but I have found that it takes, with all available methods, at least twice
as many iterations for some problems.

Thanks in advance,

Jukka Jokinen

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._