[R] lme() with known level-one variances

Fri Aug 30 20:34:45 CEST 2002

That argument is not documented in the current Windows version of nlme,
and gives an error when I try it.:

> glsControl(sigma=1)
Error in glsControl(sigma = 1) : unused argument(s) (sigma ...)
> package.description("nlme")[c("Version","Date")]
     Version         Date
    "3.1-29" "2002/08/10"

> version
         _
platform i386-pc-mingw32
arch     i386
os       mingw32
system   i386, mingw32
status
major    1
minor    5.1
year     2002
month    06
day      17
language R

R. Woodrow Setzer, Jr.                                            Phone:
(919) 541-0128
Experimental Toxicology Division                       Fax:  (919)
541-5394
Pharmacokinetics Branch
NHEERL MD-74; US EPA; RTP, NC 27711

                      "Austin, Matt"                                                                                                           
                      <maustin at amgen.co        To:       'Thomas Lumley' <tlumley at u.washington.edu>, Woodrow Setzer/RTP/USEPA/US at EPA           
                      m>                       cc:       "J.R. Lockwood" <lockwood at rand.org>, r-help at stat.math.ethz.ch                         
                                               Subject:  RE: [R] lme() with known level-one variances                                          
                      08/30/02 01:56 PM                                                                                                        

Can you do this using gls() with control = glsControl( sigma = yourValue
) ?

--Matt

-----Original Message-----
From: Thomas Lumley [mailto:tlumley at u.washington.edu]
Sent: Friday, August 30, 2002 9:12 AM
To: Setzer.Woodrow at epamail.epa.gov
Cc: J.R. Lockwood; r-help at stat.math.ethz.ch
Subject: Re: [R] lme() with known level-one variances

On Fri, 30 Aug 2002 Setzer.Woodrow at epamail.epa.gov wrote:

>
> If I understand your request correctly, you want to use something like
> "weights=varIdent(...)" as an argument to lme().  varIdent and the
other
> varFunc constructors have an argument "fixed" that allow you to
specify
> values for some or all of the coefficients of the variance function.
> See ?varIdent.  The actual error variance will be varFunc() * sigma^2,
> where sigma^2 is estimated.
>

That's the problem.

As happens in meta-analysis as well, the problem is to estimate a model
with a variance component fixed. Not fixed up to a scale parameter.
Fixed.

In meta-analysis the model is that within each trial a treatment effect
parameter is constant, and as the trial is large the variance of the
estimated treatment effect is very accurately known conditional on the
true treatment effect for that trial. The unconditional variance is then
the known conditional variance plus an unknown variance.

It doesn't seem that lme() is designed for this, and last time I tried
to
do it I gave up and changed the model more or less as you suggest.

             -thomas

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
-.-.
-.-
r-help mailing list -- Read
http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
_._.
_._

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._