s-news
[Top] [All Lists]

error from using 'weights' argument in nlme .... need help, please respo

To: s-news@wubios.wustl.edu
Subject: error from using 'weights' argument in nlme .... need help, please respond!!!!!!!
From: Savitri Appana <saviappana@yahoo.com>
Date: Fri, 29 Apr 2005 10:01:16 -0700 (PDT)
Comment: DomainKeys? See http://antispam.yahoo.com/domainkeys
Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; b=bpHDFXRMNTRc7mP6MAlFV0oSt/KRKkDrUzHC9yJzmaA6hHn220S2oU84dQfWLFBc13UCorVAdi0hK2iID3kaHsLTkfXY/slHbo5lED71B0/h9DBAWBCYN7GdjZqo257rddmj0ZxxXP+r6HShNzEwOrlXpUbeyW1MHSTQagAkcMs= ;
Hello Splus users,

I'm getting the following error when running nlme with
'weights=varFixed(~(weeks+1))'.  What does this error
mean and how can I fix this error?  Any suggestions
are much appreciated, thank you very much.

Hope to see some responses this time.

This is what i've ran and it's outcome:

> gomp.m1vf <- nlme(AVG.WT ~ Asym - (Asym - R0) * exp(
- exp(lrc) * weeks) + Asym1 *
        exp( - exp( - exp(lrc1) * (weeks - tMaxWt))), fixed =
Asym + R0 + lrc +
        Asym1 + lrc1 + tMaxWt ~ 1, random = pdDiag(Asym + R0
+ lrc + Asym1 + lrc1 +
        tMaxWt ~ 1), start = c(Asym = 339.835, R0 = -28.1426,
lrc = -2.26232, Asym1
         = 81.1841, lrc1 = -2.51447, tMaxWt = 53.1495),
weights = varFixed( ~ (
        weeks + 1)), data = geni.dat1f)
Problem in conLin$Xy * varWeights(object): length of
longer operand (15535) should be a multiple of length
of shorter (7631), while calling subroutine setup_min 

Use traceback() to see the call stack

> traceback()
23: eval(action, sys.parent())
22: doErrorAction("Problem in conLin$Xy *
varWeights(object): length of longer operand (15535)
sh
ould be a multiple of length of shorter (7631), while
calling subroutine setup_min",
21: conLin$Xy * varWeights(object)
20: recalc.varFunc(object[[i]], conLin)
19: recalc(object[[i]], conLin)
18: recalc.modelStruct(object, conLin)
17: recalc(object, conLin)
16: recalc(object, conLin)[["logLik"]]
15: logLik.lmeStruct(nlmeSt, nlmePars)
14: logLik(nlmeSt, nlmePars)
13:  - logLik(nlmeSt, nlmePars)
12: .C("setup_min",
11: ms( ~  - logLik(nlmeSt, nlmePars), start =
list(nlmePars = c(coef(nlmeSt))), 
10: NULL
9: ms( ~  - logLik(nlmeSt, nlmePars), start =
list(nlmePars = c(coef(nlmeSt))), 
8: nlme(AVG.WT ~ Asym - (Asym - R0) * exp( - exp(lrc)
* weeks) + Asym1 * exp( - exp(
7: NULL
6: nlme.formula(AVG.WT ~ Asym - (Asym - R0) * exp( -
exp(lrc) * weeks) + Asym1 * exp(
5: eval(i, local)
4: source(auto.print = auto.print, exprs =
substitute(exprs.literal))
3: script.run(exprs.literal = {
2: eval(expression(script.run(exprs.literal = {
1: 
Message: Problem in conLin$Xy * varWeights(object):
length of longer operand (15535) should be a 
multiple of length of shorter (7631), while calling
subroutine setup_min 
> 


Thank you all,
Savi


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

<Prev in Thread] Current Thread [Next in Thread>
  • error from using 'weights' argument in nlme .... need help, please respond!!!!!!!, Savitri Appana <=