Re: [R] regressing on variable with heavy tails

From: David Winsemius <dwinsemius_at_comcast.net>
Date: Sat, 30 Apr 2011 23:15:49 -0400

On Apr 30, 2011, at 9:36 PM, Jeroen Ooms wrote:

> I have a dependent variable with is very peaked and has heavy tails,
> something I haven't encountered before. (histogram:
> http://postimage.org/image/2sw9bn8pw/). What could be an appropriate
> family
> or transformation to do regress on this?

None. You have not yet established that the residuals of your analysis violate any of linear regression assumptions. It is the residuals that need to be Normal, not the dependent variable.

-- 

David Winsemius, MD
West Hartford, CT

______________________________________________
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 Sun 01 May 2011 - 04:21:20 GMT

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 Sun 01 May 2011 - 11:30:06 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.

list of date sections of archive