Re: [R] Stepwise Regression with no Origin

From: Zd Gibbs <zd.gibbs_at_yahoo.com>
Date: Mon, 18 Apr 2011 05:13:58 -0700 (PDT)

Thanks for this help. I'll read the documentation and see if I can work through the problem. I actually look forward to Mr. Harrell's reply. I have been taught that stepwise is not a good approach to use and should be avoided at all costs, so he may give me a cogent argument. The problem I posted came from my supervisor to me or I wouldn't be trying to work through the stepwise procedure. But on the brightside, it gives me an opportunity to learn something in R.

Zeda.



From: Uwe Ligges <ligges_at_statistik.tu-dortmund.de>

Cc: r-help_at_r-project.org
Sent: Fri, April 15, 2011 10:28:10 AM
Subject: Re: [R] Stepwise Regression with no Origin

On 12.04.2011 21:52, Zd Gibbs wrote:
> Sorry, the first version was incomplete. I'm trying again.
>
> I am running a regression equation and i want to enter in 12 IV then stepwise
> enter 8 variables and not have an origin.
>
> DV is "shfl".
> I  want to enter in the following 12 independent dummy variables
> ajan
> bfeb
> cmar
> dapr
> emay
> fjun
>
> And then I want to enter in a stepwise fashion
> slag6
> slag7
> slag8
> slag9
> slag10
> slag11
> slag12
>
> And finally, I want there to be no origin.
>
> I've done simple regression, but nothing quite like this.

See ?step which allows to specify minimal, maximal models as well as forward stepwise selection.

BTW: Someone called Frank Harrell will shortly let you know about possible problems in model interpretation when using stepwise variable selection.

Best,
Uwe Ligges

> I would appreciate your help.
>
> Thanks
>
> Zeda
>     [[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.

        [[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. Received on Mon 18 Apr 2011 - 12:16:31 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 Mon 18 Apr 2011 - 12:30:30 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