From: <markleeds_at_verizon.net>

Date: Thu, 22 May 2008 20:54:01 -0500 (CDT)

R-help_at_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.
Received on Fri 23 May 2008 - 02:00:27 GMT

Date: Thu, 22 May 2008 20:54:01 -0500 (CDT)

to add to what Haris said: I think you are trying to find the mle's of
the gamma, given some data, and either what you're doing is not the
right way to do it or maybe it's an alternative way to go about it that
i am unfamilar with. ?

i don't spend any time maximizing likelihoods but roger peng has a very nice example of how one can go about it using optim. i suggest that you take a look at the attached document and then write out the likelihood for the gamma similar to how i wrote it out for the poisson the other day in the private email that i sent you. in fact, the likelihood for the is probably in any decent math stat book like casella and berger etc. if you have johnson and kotz, it's definitely in there. if i had it here, i would just tell you what it was.

On Thu, May 22, 2008 at 9:36 PM, Charilaos Skiadas wrote:

> On May 22, 2008, at 9:14 PM, Edward Wijaya wrote:

*>
**>> Hi,
**>>
**>> Below I have a function mlogl_k,
**>> later it's called with "nlm" .
**>>
**>> __BEGIN__
**>> vsamples<- c(14.7, 18.8, 14, 15.9, 9.7, 12.8)
**>>
**>> mlogl_k <- function( k_func, x_func, theta_func, samp) {
**>> tot_mll <- 0
**>> for (comp in 1:k_func) {
**>> curr_mll <- (- sum(dgamma(samp, shape = x_func,
**>> scale=theta_func, log = TRUE)))
**>> tot_mll <- tot_mll + curr_mll
**>> }
**>>
**>> tot_mll
**>> }
**>>
**>> # Calling the function above
**>> mlogl_out <- nlm(mlogl_k, mean(vsamples), k_func =2, x_func = 1,
**>> theta_func = 1, samp=vsamples)
**>>
**>> __END__
**>>
**>> I thought under NLM, I already assign
**>> the parameter correctly.
**>> However it gives me the following error.
**>>
**>> Error in f(x, ...) : unused argument(s) (14.3166666666667)
**>> Calls: nlm -> <Anonymous> -> f
**>> Execution halted
**>>
**>>
**>> What's wrong with my code above?
**>
**> nlm passes the following values over to mlogl_k: mean(vsamples),
**> k_func =2, x_func = 1, theta_func = 1, samp=vsamples
**>
**> Now, f is called with these arguments, and all its arguments are
**> matched by the named arguments. Then it does not know what to do with
**> the "mean(vsamples)" argument. Hence the error: f is called with one
**> more argument than it can handle.
**>
**> It is not clear with respect to what variables you want the
**> minimization. Right now all the named variables are treated as
**> constants, and there is no variable left with respect to which mlogl_k
**> is supposed to be maximized. Read the documentation for nlm carefully:
**> Your function should have one more argument (like the alpha in the
**> question you asked yesterday).
**>
**>> - Edward
**>
**> Haris Skiadas
**> Department of Mathematics and Computer Science
**> Hanover College
**>
**> ______________________________________________
**> R-help_at_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.
*

R-help_at_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.

- application/pdf attachment: r-classes-scope.pdf

Archive maintained by Robert King, hosted by
the discipline of
statistics at the
University of Newcastle,
Australia.

Archive generated by hypermail 2.2.0, at Fri 23 May 2008 - 03:30:41 GMT.

*
Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help.
Please read the posting
guide before posting to the list.
*