<html><body><div style="color:#000; background-color:#fff; font-family:lucida console, sans-serif;font-size:12pt"><div style="font-family: 'lucida console', sans-serif; font-size: 12pt; ">Desculpe Walmes por não ter mandado o CMR antes!</div><div style="font-family: 'lucida console', sans-serif; font-size: 12pt; "><br></div><div><div style="font-family: 'lucida console', sans-serif; font-size: 12pt; color: rgb(69, 69, 69); "><font color="#454545" face="times new roman, new york, times, serif">Bom, vamos lá de novo! </font><span style="font-family: 'times new roman', 'new york', times, serif; font-size: 12pt; ">Pessoal bom dia!</span></div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; "><br></div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; ">Seguinte, estou trabalhando com modelos não lineares mistos. Ajustei alguns
modelos e quando fui sugerir uma estrutura para a matriz de variâncias e covariâncias intra-indivíduos (R), com a estrutura AR com heterogeneidade (ARH) para esta matriz, o modelo rodou, mas soltou a seguinte msn:<br><br></div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; ">"Warning message:<br>Singular precision matrix in level -1, block 1"<br> <br>Gostaria de perguntar p vcs o que essa msn pode influenciar nos resultados?</div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; "><br></div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; ">Pq ao fazer o teste da razao de verossimilhanças, este modelo com estrutura ARH foi selecionado, melhorou os resíduos e a normalidade deste, mas ao desenhar as curvas o ajuste me pareceu não tão satisfatório como aquele
q eu tinha feito considerando R = I * \sigma^2.</div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; ">Alguém tem alguma luz ai?</div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; "><br></div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; ">A seguir um CMR e os dados (http://www.datafilehost.com/download-a20803e9.html):</div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; "><br></div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; "><span style="font-size: medium; ">#################################################################################</span> <br></div><div style="background-color: transparent; "><div
style="background-color: transparent; "><font color="#454545" face="times, serif">dados<- read.table('dados.txt', head=T)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">names(dados)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif"><br></font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">tempo <- dados$Tempo</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">periodo<- factor(dados$Periodo)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">animal <- factor(dados$Animal)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">IND <- dados$Rep</font></div><div style="background-color: transparent; "><font color="#454545" face="times,
serif">trat <- factor(dados$Tratamento)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">FDN <- dados$FDN; length(FDN)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">Dados_FDN<- na.omit(data.frame(tempo, periodo, animal, IND, trat, FDN))</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif"><br></font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">TRAT3<- factor(c(rep(70,120), rep(50,120), rep(30,118)))</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">DADOS3<- cbind(Dados_FDN,TRAT3)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif"><br></font></div><div style="background-color: transparent; "><font color="#454545" face="times,
serif"># Função</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">fx = function(t,a,b,c) (a + b*(1-exp(-c*t)))</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif"># chutes</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">a1=1.13000000; b1=72.04000000; c1=0.02501543</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif"><br></font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">library(lattice)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">options( contrasts = c("contr.treatment", "contr.poly") )</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">nlmeControl(500)</font></div><div
style="background-color: transparent; "><font color="#454545" face="times, serif"><br></font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">require(nlme)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">fdn3=groupedData(FDN~tempo|IND,data=DADOS3, outer=~factor(TRAT3), order.groups=F)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif"><br></font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">FDN3.list = nlsList(FDN ~ fx(tempo,a,b,c)|IND, data=fdn3, start=list(a=a1, b=b1, c=c1))</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">FDN.30<- nlme(FDN3.list, fixed=a+b+c~TRAT3)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">abcfix3 = fixef(FDN.30)</font></div><div
style="background-color: transparent; "><font color="#454545" face="times, serif"><br></font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">FDNab_ARH <- update(FDN.30, random = pdSymm(a+b+c ~ 1 ),</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif"> fixed = list(a+b~1,c~TRAT3), start=c(abcfix3[1],</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif"> abcfix3[2],</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">
abcfix3[3],0,0),</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif"> corr = corAR1(form=~1|IND), weight=varIdent(form=~1|tempo))</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">summary(FDNab_ARH)</font></div><div style="background-color: transparent; "><font color="#454545" face="times, serif">#################################################################################</font></div></div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; "><br></div><div style="font-family: times, serif;
font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; ">desde já muito obrigada!</div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; ">Simone</div><div style="font-family: times, serif; font-size: 12pt; color: rgb(69, 69, 69); background-color: transparent; "><br></div></div> </div></body></html>