windows XP
R 2.10
As pointed out by Prof. Venables and Ripley (MASS 4th edition, p275), the results obtained from lme using method="ML" and method="REML" are often different, especially for small datasets. Is there any way to determine which method is preferable for a given set of data?
Thanks,
john
John David Sorkin M.D., Ph.D.
Chief, Biostatistics and Informatics
University of Maryland School of Medicine Division of Gerontology
Baltimore VA Medical Center
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524
(Phone) 410-605-7119
(Fax) 410-605-7913 (Please call phone number above prior to faxing)
Confidentiality Statement:
This email message, including any attachments, is for th...{{dropped:6}}
lm: RME vs. ML
3 messages · John Sorkin, JLucke at ria.buffalo.edu, Peter Dalgaard
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20091208/76880665/attachment-0001.pl>
JLucke at ria.buffalo.edu wrote:
You need to give your criteria for "preferable". For normal-linear models, REML estimates of variances are unbiased, whereas ML estimates are downwardly biased.
I suspect that you can't actually say anything general about the direction of the bias, except for the residual term. In the cases that can be analyzed explicitly, variance estimates are complicated linear combinations of sums of squares, and if those are not biased by the same relative amount, the net result could conceivably be a positive bias. I could be wrong though.
My intuition is that the ML-induced bias would be worse in small samples. I don't know about other distributions. Likewise I don't know about MSE or other criterion for preference.
O__ ---- Peter Dalgaard ?ster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk) FAX: (+45) 35327907