[R] Re : standard error of logit parameters

[markleeds at verizon.net]

I'm sure below is fine but  john fox's CAR book has some nice examples 
of how to compute the logit parameters and variances from scratch using 
iteratively weighted least squares.



On Thu, Jan 29, 2009 at  1:54 AM, justin bem wrote:

> Run
>
> outfit<-nlm(..., hessian=T) and then standards error are
> se<-diag(solve(outfit$hessian))
>
>
> ��
> Justin BEM
> BP 1917 Yaound��
> T��l (237) 76043774
>
> ��
>
>
>
> ________________________________
> De : Bomee Park <bombom at stanford.edu>
> �� : r-help at r-project.org
> Envoy�� le : Jeudi, 29 Janvier 2009, 4h01mn 56s
> Objet��: [R] standard error of logit parameters
>
> Hi everyone.
>
> I am now estimating the parameters for a logit model, and trying to 
> get the estimates by laximizing the log_likelihood.
> The nlm function works nicely for maximizing the -(log_likelihood) and 
> returns the parameter estimates that minimize the static, and the 
> gradients also, but don't have any clue how I can get the standard 
> error for the parameters.
>
> Any help will be greatly appreciated.
> Thanks.
>
> ______________________________________________
> 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.
>
>
>       	[[alternative HTML version deleted]]
>
>
>
>      ------------------------------
>
> ______________________________________________
> 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.