Wouter,
> what is the best way to do an ANOVA on two-factor experiments with
> repeated measures on one of the factors (e.g. time) in R, (with
> Greenhouse-Geisser Epsilon or Huynh-Feldt Epsilon calculation, if
You could use aov() with an Error term on your subjects factor; this follows
the classical "split-plot" approach. But I do not believe you will
find any
existing R code to do the G-G or H-F corrections. And I doubt there ever
will be, because the better strategy is to use linear mixed-effect modeling
approaches. See for example, Diggle et al (Analysis of Longitudinal Data,
2001); Pinheiro & Bates (Mixed Effects Model with S, 2000); and Venables
&
Ripley (MASS 4th edition, Sitka Trees example.) You get the split-plot
approach with a uniform correlation (also called compound symmetry)
specification for your error matrix, but other alternatives are available.
If your repeated measure is time, my experience (with biology data) finds
that a first order autoregression or something similar is better. The
correlation between timepoint pairs weakens as the distance between them
increases.
Hope that helps.
Best Regards,
Bill
----------------------------------------
Bill Pikounis, Ph.D.
Biometrics Research Department
Merck Research Laboratories
PO Box 2000, MailDrop RY33-300
126 E. Lincoln Avenue
Rahway, New Jersey 07065-0900
USA
v_bill_pikounis at merck.com
Phone: 732 594 3913
Fax: 732 594 1565
> -----Original Message-----
> From: wouter buytaert [mailto:wouter.buytaert at yucom.be]
> Sent: Friday, July 18, 2003 6:34 AM
> To: R-help at stat.math.ethz.ch
> Subject: [R] repeated measures anova
>
>
> Hello,
>
> what is the best way to do an ANOVA on two-factor experiments with
> repeated measures on one of the factors (e.g. time) in R, (with
> Greenhouse-Geisser Epsilon or Huynh-Feldt Epsilon calculation, if
> possible (as described in Ott and Longnecker, 2001.
> Statistical Methods
> and Data Analysis 5th ed., chapter 18))
>
> Thanks,
>
> Wouter
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
>
------------------------------------------------------------------------------
Notice: This e-mail message, together with any attachments, ...{{dropped}}