Re: [Rd] nobs() and logLik()

From: Michael Friendly <friendly_at_yorku.ca>
Date: Fri, 20 Jan 2012 09:19:30 -0500

On 1/20/2012 1:42 AM, Berwin A Turlach wrote:
> One reason that I can see for people to use zero weights rather than
> 'subset' is that fitted() and predict() in the former case readily
> produce fitted values for the observations that received a zero weight.
>
Another is that including the case of zero weights naturally allows a variety of simple robust methods via a weight function that descends to 0. A discontinuity at 0 in the handling of weights prevents this use.

-- 
Michael Friendly     Email: friendly AT yorku DOT ca
Professor, Psychology Dept.
York University      Voice: 416 736-5115 x66249 Fax: 416 736-5814
4700 Keele Street    Web:   http://www.datavis.ca
Toronto, ONT  M3J 1P3 CANADA

______________________________________________
R-devel_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Received on Fri 20 Jan 2012 - 14:21:49 GMT

This quarter's messages: by month, or sorted: [ by date ] [ by thread ] [ by subject ] [ by author ]

All messages

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 Sat 21 Jan 2012 - 19:00:10 GMT.

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

list of date sections of archive