R help - Jun 2010 - xyplot: adding pooled regression lines to a paneled type="r" plot

Consider the following plot that shows separate regression lines ~ age 
for each subject in the Pothoff-Roy Orthodont data,
with separate panels by Sex:

library(nlme)
#plot(Orthodont)
xyplot(distance ~ age|Sex, data=Orthodont, type='r', groups=Subject, 
col=gray(.50),
    main="Individual linear regressions ~ age")

I'd like to also show in each panel the pooled OLS regression line for 
each Sex in the corresponding panel,
generated by the following model:

Ortho.OLS <- lm(distance ~ age * Sex, data=Orthodont)

Sex is a factor, with Male=0, so the coefficients are:
 > coef(Ortho.OLS)
  (Intercept)           age     SexFemale age:SexFemale
   16.3406250     0.7843750     1.0321023    -0.3048295

I anticipate wanting to fit other models to these data, and also 
displaying the model-predicted
regression lines in the same or similar plot, e.g., for a simple linear 
mixed model:

Ortho.MLM <- lme(distance ~ age * Sex, data=Orthodont,
        random = ~ 1 + age | Subject,
        correlation = corAR1 (form = ~ 1 | Subject))

TIA,
-Michael


-- 
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

Michael,

Look at the ancova function in the HH package.

## install.packages("HH")  ## if not there yet.
library(HH)
library(nlme)  ## for the Orthodont data
ancova(distance ~ age*Sex, data=Orthodont)
ancova(distance ~ age+Sex, data=Orthodont)
ancova(distance ~ age, groups=Sex, data=Orthodont)
ancova(distance ~ Sex, x=age, data=Orthodont)

Rich


On Tue, Jun 22, 2010 at 12:30 PM, Michael Friendly
<friendly@yorku.ca>wrote:
> Consider the following plot that shows separate regression lines ~ age for
> each subject in the Pothoff-Roy Orthodont data,
> with separate panels by Sex:
>
> library(nlme)
> #plot(Orthodont)
> xyplot(distance ~ age|Sex, data=Orthodont, type='r',
groups=Subject,
> col=gray(.50),
>   main="Individual linear regressions ~ age")
>
> I'd like to also show in each panel the pooled OLS regression line for
each
> Sex in the corresponding panel,
> generated by the following model:
>
> Ortho.OLS <- lm(distance ~ age * Sex, data=Orthodont)
>
	[[alternative HTML version deleted]]

On Tue, Jun 22, 2010 at 9:30 AM, Michael Friendly <friendly at yorku.ca>
wrote:
> Consider the following plot that shows separate regression lines ~ age for
> each subject in the Pothoff-Roy Orthodont data,
> with separate panels by Sex:
>
> library(nlme)
> #plot(Orthodont)
> xyplot(distance ~ age|Sex, data=Orthodont, type='r',
groups=Subject,
> col=gray(.50),
> ? main="Individual linear regressions ~ age")
>
> I'd like to also show in each panel the pooled OLS regression line for
each
> Sex in the corresponding panel,
> generated by the following model:
>
> Ortho.OLS <- lm(distance ~ age * Sex, data=Orthodont)
>
> Sex is a factor, with Male=0, so the coefficients are:
>> coef(Ortho.OLS)
> ?(Intercept) ? ? ? ? ? age ? ? SexFemale age:SexFemale
> ?16.3406250 ? ? 0.7843750 ? ? 1.0321023 ? ?-0.3048295
>
> I anticipate wanting to fit other models to these data, and also displaying
> the model-predicted
> regression lines in the same or similar plot, e.g., for a simple linear
> mixed model:
>
> Ortho.MLM <- lme(distance ~ age * Sex, data=Orthodont,
> ? ? ? random = ~ 1 + age | Subject,
> ? ? ? correlation = corAR1 (form = ~ 1 | Subject))
Have a look at

http://user2007.org/program/presentations/sarkar.pdf

-Deepayan