[R] lattice add a fit

[Bert Gunter]

You need to explicitly pass the subscripts arguments and your
covariates. See ?xyplot for details. Your panel function call would
then be something like:

        panel=function(x,y,subscripts,w,z, ...){
            panel.xyplot(x,y,...)
             panel.loess(x,y,...,col.line="red"
            mod <- lm(y~x + w + z, data = data[subscripts,])
           ... etc.
}


Cheers,
Bert



Bert Gunter
Genentech Nonclinical Biostatistics
(650) 467-7374

"Data is not information. Information is not knowledge. And knowledge
is certainly not wisdom."
Clifford Stoll




On Tue, Oct 7, 2014 at 8:13 AM, Bond, Stephen <Stephen.Bond at cibc.com> wrote:
> Bert,
>
> Can you provide an example how to pass the conditioned data set?
>
> xyplot( x~y | z, data=mydata,
> panel=function(...){
> mod=lm(x~ y+w +q, data=??)
> panel.lines(fitted(mod))
>
> }
>
> If I use mydata in place of ?? I get a global fit, not a fit for each level of z, which is what I want.
>
> Stephen B
>
>
> -----Original Message-----
> From: Bert Gunter [mailto:gunter.berton at gene.com]
> Sent: Tuesday, October 07, 2014 9:30 AM
> To: Bond, Stephen
> Cc: r-help at R-project.org
> Subject: Re: [R] lattice add a fit
>
> Fit your model in the panel function using lm and plot the fits using ?panel.points, ?panel.lines, etc.
>
> -- Bert
>
> Bert Gunter
> Genentech Nonclinical Biostatistics
> (650) 467-7374
>
> "Data is not information. Information is not knowledge. And knowledge is certainly not wisdom."
> Clifford Stoll
>
>
>
>
> On Tue, Oct 7, 2014 at 6:01 AM, Bond, Stephen <Stephen.Bond at cibc.com> wrote:
>> What is the way to add an arbitrary fit from a model to a lattice conditioning plot ?
>>
>> For example
>> xyplot(v1 ~v2 | v3,data=mydata,
>>         panel=function(...){
>>             panel.xyplot(...)
>>             panel.loess(...,col.line="red")
>>         }
>> )
>> Will add a loess smoother. Instead, I want to put a fit from lm (but
>> not a simple straight line) and the fit has to be done for each panel separately, not one fit for the full data set, so sth like an lm equivalent of panel.locfit (there is no panel.lmfit) Thank you.
>>
>> Stephen B
>>
>>
>>         [[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.