Re: [R] Fwd: Plotting confidence intervals for lme

From: Spencer Graves <spencer.graves_at_pdf.com>
Date: Fri 08 Jul 2005 - 12:46:01 EST

          Have you considered the examples with "?TukeyHDS"? This may not work with a standard "lme" object. However, I suspect that I could probably get something useful from this for an "lme" by walking through the code line by line.

          spencer graves

Ghislain Vieilledent wrote:

> ---------- Forwarded message ----------
> From: Ghislain Vieilledent <ghislainv@gmail.com>
> Date: 7 juil. 2005 11:38
> Subject: Re: [R] Plotting confidence intervals for lme
> To: Spencer Graves <spencer.graves@pdf.com>
> 
> That's what I was looking for. Thanks a lot!
> 
> That's true units are mixing and that values of coef depend on the contrasts 
> but I want to use this graph in order to compare levels' coefficient of each 
> of the factor taken independantly, not to compare factors' coefficients to 
> eachothers. Does it mean something?
> 
> Thanks again. 
> 
> Ghislain Vieilledent.
> 
> 
> 
> 2005/7/6, Spencer Graves <spencer.graves@pdf.com>:
> 

>>Consider the following extension of an example in "?lme":
>>
>>fm2 <- lme(distance ~ age + Sex, data = Orthodont, random = ~ 1)
>>int <- intervals(fm2)
>>class(int$fixed)
>>kf <- dim(int$fixed)[1]
>>plot(int$fixed[,2], kf:1,
>>xlab="x", ylab="", xlim=range(int$fixed),
>>axes=FALSE)
>>axis(1)
>>axis(2, kf:1, dimnames(int$fixed)[[1]])
>>segments(int$fixed[,1], kf:1,
>>int$fixed[,3], kf:1)
>>abline(v=0)
>>
>>However, we are mixing units, i.e., the units for the intercept are
>>"distance", while for "age" are "distance/time", and the interpretation
>>of the coefficient of a factor like Sex depends on contrasts used.
>>Thus, I don't know how much sense it makes to prepare plots like this.
>>
>>For similar plots that make more sense, see Pinheiro and Bates (2000
>>Mixed-Effects Models in S and S-PLUS (Springer).
>>
>>spencer graves
>>
>>Ghislain Vieilledent wrote:
>>
>>
>>>Hello and sorry to disturb.
>>>
>>>I'm trying to plot the confidence intervals for the fixed effects of a
>>
>>lme.
>>
>>>I want to obtain graphically, if it is possible, a bar with Estimate,
>>
>>upper
>>
>>>and lower CI for each level of the factors.
>>>
>>>I know how to do for a lm model but for a lme one, I tried with
>>>plot(intervals(...)) and plot(ci(...)) from the gmodels package but it
>>>doesn't work well.
>>>
>>>Thanks for you help and have a good day.
>>>
>>
>>--
>>Spencer Graves, PhD
>>Senior Development Engineer
>>PDF Solutions, Inc.
>>333 West San Carlos Street Suite 700
>>San Jose, CA 95110, USA
>>
>>spencer.graves@pdf.com
>>www.pdf.com <http://www.pdf.com> <http://www.pdf.com>
>>Tel: 408-938-4420
>>Fax: 408-280-7915
>>
> 
> 
> 
> 

-- 
Spencer Graves, PhD
Senior Development Engineer
PDF Solutions, Inc.
333 West San Carlos Street Suite 700
San Jose, CA 95110, USA

spencer.graves@pdf.com
www.pdf.com <http://www.pdf.com>
Tel:  408-938-4420
Fax: 408-280-7915

______________________________________________
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Received on Fri Jul 08 12:50:52 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:33:22 EST