Dear Brian/Arthur,
I have a dataset (ecoli.txt) which contains the following information.
• 3 trials (trial1, trial2 and trial3)
• 2 treatments (trt1 vs trt2)
• Different sampling dates(sdate) were recorded for each treatment in each trial
trial1: 7, 14, 20, 34, 56, 69, 83, 112, 154, 181, 205
trial2: 14, 27, 48, 69, 97, 125, 153, 183
trial3: 14, 27, 55, 76, 97, 133, 153, 174
• Within each trial for each sampling date, logcount of ecoli were measured from sampling 3 different plates of 2 or 3 chambers of 3 blocks (A, B, C).
trial1: 3 chambers /block
trials 2 and 3: 2 chambers/block.
The aim of the analysis is to examine the difference between two treatments in each trial and across trials, and predict the relationship between sampling date (sdate) and logcount for each treatment.
I have used ASReml_R to fit the following model:
asreml(logcount~trial+trial/treatment+trial/(sdate+sd2)+trial/treatment:(sdate+sd2),data=ecoli,random=~(trial/sdate):block/chamber/plate)
The term sd2 is a quadratic of covariate sdate.
The program gives the results, but it does not give a denominator df when I ask for conditional wald tests.
Is my model wrong?
The ASreml_R code:
rm(list=ls())
ecoli=read.table("C:\\Documents and Settings\\li042\\Desktop\\ecoli.txt",sep="\t",header=T)
library(asreml)
ecoli$chamber <- as.factor(ecoli$chamber)
ecoli$plate <- as.factor(ecoli$plate)
ecoli$sd2<-ecoli$sdate^2
ecoli_m2.asr <- asreml(logcount~trial+trial/treatment+trial/(sdate+sd2)+trial/(treatment:(sdate+sd2)),data=ecoli,random=~(trial/sdate):block/chamber/plate)
wald.asreml(ecoli_m2.asr, denDF = "default", ssType = "conditional")$Wald
The data file is attached. Thanks in advance.
Yutao Li
-------------------- m2f --------------------
Sent using Mail2Forum (http://www.mail2forum.com).
Read this topic online here:
http://www.vsni.co.uk/forum/viewtopic.php?p=802#802
-------------------- m2f --------------------
Attachments:
http://www.vsni.co.uk/forum/files/ecoli_633.txt
Received on Fri Sep 03 2009 - 04:27:15 EST
This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)