Boa tarde,
Estou tentando modelar as funções gompertz e logístico porem quero somar as
duas equações e gerar um único gráfico. Porem a equação terá 6 parâmetros teria
alguma rotina para esse caso. Irei anexar a rotina utilizada quando uma equação
é modelada de cada vez e os dados MFF que eu utilizei para modelar essas
funções. Meu novo arquivo de dados GG (y são dois caracteres juntos) por isso
que na rotina teria que somar a equação de gompertz (y~b0*exp(-exp(b1-b2*x)
porem quando somo essa equação com a mesma equação de gompertz da erro. Eu
terei 6 parâmetros nesse caso , já tentei indicar de varias formas o b0, b1 e
b2 pois além desses tenho mais um b0, b1 e b2. Espero que eu tenha me feito
entender.
Att Cláudia de Bem
________________________________
De: R-br <r-br-boun...@listas.c3sl.ufpr.br> em nome de ASANTOS por (R-br)
<r-br@listas.c3sl.ufpr.br>
Enviado: terça-feira, 29 de janeiro de 2019 18:45
Para: r-br@listas.c3sl.ufpr.br; r-br-requ...@listas.c3sl.ufpr.br
Cc: ASANTOS
Assunto: [R-br] Ajuda para plotar IC para modelo glm com junção de níveis
Prezados Membros do R-br,
Estou tentando plotar um modelo glmcom stat_smooth() no ggplot2 sem
sucesso. Minha ideia é representar o intervalo de confiança para um
modelo onde realizei a junção de dois níveis para o fator Feature. Ou
seja, modelos iguais são representados por uma única curva, porém a
função stat_smooth() não "enxerga" os níveis que foram juntados. Segundo
exemplo fictício do meu problema:
#Simulação de um banco de dados, onde a variável resposta Production tem
distribuição Gamma
set.seed(123)
d<-NULL
N<-50
d$Production <- rgamma(N,10)
d$Feature <- ifelse(d$Production >7 & d$Production<10,
c("green"),ifelse(d$Production>11,
c("red"), c("blue")))
d$Temp<-rnorm(N,20,5)
d<-as.data.frame(d)
#
# Ajusto um modelo de Gamma completo
mG<- glm(Production~ Feature + Temp, family= Gamma, data = d)
summary(mG)
anova(mG,test="Chi")
# Comparo os níveis do fator Feature
library(multcomp)
PW<-summary(glht(mG, linfct = mcp(Feature= "Tukey")))
PW
cld(PW)
# Faço de conta que os níveis green = blue e realizo um novo ajuste
Feature2<-d$Feature
levels(Feature2)
levels(Feature2)[1]<-"blue&green"
levels(Feature2)[2]<-"blue&green"
levels(Feature2)
d$Feature2<-Feature2
mG2<- glm(Production~ Feature2 + Temp, family= Gamma, data = d)
# Crio os valores de predição
pred.data = data.frame(
Feature2<-d$Feature2,
Temp<-d$Temp
)
pred.data$Production = predict(mG2, newdata=pred.data, type="response")
#Realizo o plot
library("ggplot2")
ggplot(d, aes(Temp, Production, colour = Feature2)) +
geom_point() +
geom_line(data=pred.data) +
stat_smooth(method = "glm", formula = y ~ x, family = Gamma)
#
E tenho uma representação errada do intervalo de confiança do meu
modelo mG2 porque não consigo fazer a função stat_smooth() considerar o
ajuste mG2. Como posso resolver isso?
Obrigado
--
======================================================================
Alexandre dos Santos
Proteção Florestal
IFMT - Instituto Federal de Educação, Ciência e Tecnologia de Mato Grosso
Campus Cáceres
Caixa Postal 244
Avenida dos Ramires, s/n
Bairro: Distrito Industrial
Cáceres - MT CEP: 78.200-000
Fone: (+55) 65 99686-6970 (VIVO) (+55) 65 3221-2674 (FIXO)
e-mails:alexandresanto...@yahoo.com.br
alexandre.san...@cas.ifmt.edu.br
Lattes: http://lattes.cnpq.br/1360403201088680
OrcID: orcid.org/0000-0001-8232-6722 - ResearcherID: A-5790-2016
Researchgate:
www.researchgate.net/profile/Alexandre_Santos10<http://www.researchgate.net/profile/Alexandre_Santos10>
LinkedIn: br.linkedin.com/in/alexandre-dos-santos-87961635
Mendeley:www.mendeley.com/profiles/alexandre-dos-santos6/
======================================================================
_______________________________________________
R-br mailing list
R-br@listas.c3sl.ufpr.br
https://listas.inf.ufpr.br/cgi-bin/mailman/listinfo/r-br
Leia o guia de postagem (http://www.leg.ufpr.br/r-br-guia) e fornea cdigo mnimo
reproduzvel.
#########################################################################
#### ####
#### Rotina Modelos Nao-Lineares ####
#### ####
####-----------------------------------------------------------------####
#### ####
#### + Logistico ####
#### + Gompertz ####
#### ####
#########################################################################
#########################################################################
#### Definicao Modelo ####
#########################################################################
#### Logistico
Logistico <- function(x,b0,b1,b2){b0/(1+exp(b1-b2*x))}
#### Gompertz
Gompertz <- function(x,b0,b1,b2){b0*exp(-exp(b1-b2*x))}
#########################################################################
#### Load Packages ####
#########################################################################
library(MASS)
library(lmtest)
library(car)
library(nortest)
library(manipulate) ## Permite criar gráficos "móveis"
library(readxl)
#########################################################################
#### Load Dados ####
#########################################################################
dados <- read.table("C:/Resultados R pos doc/MFF.txt",h=T, dec=".")
str(dados)
head(dados)
summary(dados)
###Determinação das estimativas iniciais, para cada modelo.
#########################################################################
#### Estimarivas Iniciais ####24
#########################################################################
#### Logistico
b0=200 #Assintota com chute inicial "bem errado"
b1=3.5 #Parâmetro de localização com chute inicial "bem errado"
b2=0.005 #Taxa de crescimento com chute inicial "bem errado"
plot(y~x,data=dados)
curve(Logistico(x,b0,b1,b2),add=T,col=2)
Chute_logist <- list() ## Lista em branco onde serão armazenados os chutes
### Programação específica do gráfico "movel"
manipulate({
plot(y~x,data=dados)
curve(Logistico(x,b0=b0,b1=b1,b2=b2),add=TRUE)
Chute_logist <<-list(b0=b0,b1=b1,b2=b2)},
b0=slider(0,350,initial=0),
b1=slider(0,6.1,initial=0),
b2=slider(0.001,0.9,initial=0.001)
)
Chute_logist ## onde estão armazenados os chutes
#### Gompertz
b0=200 #Assintota com chute inicial "bem errado"
b1=3.5 #Parâmetro de localização com chute inicial "bem errado"
b2=0.005 #Taxa de crescimento com chute inicial "bem errado"
plot(y~x,data=dados)
curve(Gompertz(x,b0,b1,b2),add=T,col=2)
Chute_gomp <- list() ## Lista em branco onde serão armazenados os chutes
### Programação específica do gráfico "movel"
manipulate({
plot(y~x,data=dados)
curve(Gompertz(x,b0=b0,b1=b1,b2=b2),add=TRUE)
Chute_gomp <<-list(b0=b0,b1=b1,b2=b2)},
b0=slider(0,350,initial=0),
b1=slider(0,6.1,initial=0),
b2=slider(0.001,0.9,initial=0.001))
Chute_gomp ## onde estão armazenados os chutes
#########################################################################
#### Ajuste Modelo ####
#########################################################################
#### Logistico
Modelo_Logist <- nls(y~b0/(1+exp(b1-b2*x)), data=dados,start=Chute_logist)
ResumoModel_Logist <- summary(Modelo_Logist)
b0_logist <- summary(Modelo_Logist)$parameters[1,1]
b1_logist <- summary(Modelo_Logist)$parameters[2,1]
b2_logist <- summary(Modelo_Logist)$parameters[3,1]
IC_logist <- confint.default(Modelo_Logist)
#### Gompertz
Modelo_Gompertz <- nls(y~b0*exp(-exp(b1-b2*x)), data=dados,start=Chute_gomp)
ResumoModel_Gompertz <- summary(Modelo_Gompertz)
b0_gomp <- summary(Modelo_Gompertz)$parameters[1,1]
b1_gomp <- summary(Modelo_Gompertz)$parameters[2,1]
b2_gomp <- summary(Modelo_Gompertz)$parameters[3,1]
IC_gomp <- confint.default(Modelo_Gompertz, level = .95)
#########################################################################
#### Pressupostos Modelo ####
#########################################################################
#### Logitico
# Normalidade
SW_logist <- shapiro.test(residuals(Modelo_Logist))
LL_logist <- lillie.test(residuals(Modelo_Logist))
KS_logist <- ks.test(residuals(Modelo_Logist), "pnorm",
mean(residuals(Modelo_Logist)), sd(residuals(Modelo_Logist)))
AD_logist <- ad.test(residuals(Modelo_Logist))
# Homogeneidade de variacia
gradiente_mod_logist <- attr(Modelo_Logist$m$fitted(),"gradient") # obtem
matriz gradiente, sobre a qual será feito o diagnóstico
m0_mod_logist <- lm(y~-1+gradiente_mod_logist, data=dados) # passando a m atriz
gradiente para a lm(), importante remover intercepto (-1)
BP_logist <- bptest(m0_mod_logist) # teste de Breusch-Pagan (homogeneidade)
BP_logist
BAR_logist <- bartlett.test(residuals(Modelo_Logist)~x,data=dados)
BAR_logist
# Idependencia
DW_logist <- durbinWatsonTest(m0_mod_logist) # teste de DW (independência)
DW_logist
# Não linearidade
## Função deriv3 retorna a matriz hessiana necessária para calcular a não
linearidade
## pelo método de Bates e Watts
prmt_mod_logist <- deriv3(~b0/(1+exp(b1-b2*x)),c("b0","b1","b2"),function
(x,b0,b1,b2)NULL)
nls_logist <- nls(y~prmt_mod_logist(x,b0,b1,b2),data=dados, start=Chute_logist)
rms.curv(nls_logist)
parm_R1_logist <- rms.curv(nls_logist)$pe
intrinseca_R1_logist <- rms.curv(nls_logist)$ic
#### Gompertz
# Normalidade
####### Analisando os resíduos
SW_gomp <- shapiro.test(residuals(Modelo_Gompertz))
LL_gomp <- lillie.test(residuals(Modelo_Gompertz))
KS_gomp <- ks.test(residuals(Modelo_Gompertz), "pnorm",
mean(residuals(Modelo_Gompertz)), sd(residuals(Modelo_Gompertz)))
AD_gomp <- ad.test(residuals(Modelo_Gompertz))
# Homogeneidade
gradiente_mod_gomp <- attr(Modelo_Gompertz$m$fitted(),"gradient") # obtem
matriz gradiente, sobre a qual será feito o diagnóstico
m0_mod_gomp <- lm(y~-1+gradiente_mod_gomp, data=dados) # passando a matriz
gradiente para a lm(), importante remover intercepto (-1)
BP_gomp <- bptest(m0_mod_gomp) # teste de Breusch-Pagan (homogeneidade)
BP_gomp
BAR_gomp <- bartlett.test(residuals(Modelo_Gompertz)~x,data=dados)
BAR_gomp
# Independencia
DW_gomp <- durbinWatsonTest(m0_mod_gomp) # teste de DW (independência)
DW_gomp
# Não linearidade
## Função deriv3 retorna a matriz hessiana necessária para calcular a não
linearidade
## pelo método de Bates e Watts
prmt_mod_gomp <- deriv3(~b0*exp(-exp(b1-b2*x)),c("b0","b1","b2"),function
(x,b0,b1,b2)NULL)
nls_gomp <- nls(y~prmt_mod_gomp (x,b0,b1,b2),data=dados, start=Chute_gomp)
rms.curv(nls_gomp)
parm_R1_gomp <- rms.curv(nls_gomp)$pe
intrinseca_R1_gomp <- rms.curv(nls_gomp)$ic
#########################################################################
#### Pontos Criticos ####
#########################################################################
# Logistico
PIX_logist <-
summary(Modelo_Logist)$parameters[2,1]/summary(Modelo_Logist)$parameters[3,1]
PIY_logist <- summary(Modelo_Logist)$parameters[1,1]/2
PAMX_logist <-
(summary(Modelo_Logist)$parameters[2,1]-log(2+sqrt(3)))/summary(Modelo_Logist)$parameters[3,1]
PAMY_logist <- summary(Modelo_Logist)$parameters[1,1]/(3+sqrt(3))
PDMX_logist <-
(summary(Modelo_Logist)$parameters[2,1]-log(2-sqrt(3)))/summary(Modelo_Logist)$parameters[3,1]
PDMY_logist <- summary(Modelo_Logist)$parameters[1,1]/(3-sqrt(3))
PDAX_logist <-
(summary(Modelo_Logist)$parameters[2,1]-log(5-2*sqrt(6)))/summary(Modelo_Logist)$parameters[3,1]
PDAY_logist <- summary(Modelo_Logist)$parameters[1,1]/(2*(3-sqrt(6)))
# Gompertz
PIX_gomp <-
summary(Modelo_Gompertz)$parameters[2,1]/summary(Modelo_Gompertz)$parameters[3,1]
PIY_gomp <- summary(Modelo_Gompertz)$parameters[1,1]/2
PAMX_gomp <-
(summary(Modelo_Gompertz)$parameters[2,1]-log(2+sqrt(3)))/summary(Modelo_Gompertz)$parameters[3,1]
PAMY_gomp <- summary(Modelo_Gompertz)$parameters[1,1]/(3+sqrt(3))
PDMX_gomp <-
(summary(Modelo_Gompertz)$parameters[2,1]-log(2-sqrt(3)))/summary(Modelo_Gompertz)$parameters[3,1]
PDMY_gomp <- summary(Modelo_Gompertz)$parameters[1,1]/(3-sqrt(3))
PDAX_gomp <-
(summary(Modelo_Gompertz)$parameters[2,1]-log(5-2*sqrt(6)))/summary(Modelo_Gompertz)$parameters[3,1]
PDAY_gomp <- summary(Modelo_Gompertz)$parameters[1,1]/(2*(3-sqrt(6)))
#########################################################################
#### Qualidade Modelo ####
#########################################################################
# Logistico
SQR_logist <- sum((dados$y-fitted(Modelo_Logist))^2) ; SQR_logist
SQT_logist <- sum(dados$y*dados$y) - (sum(dados$y)^2)/length(dados$y) ;
SQT_logist
R2_logist <- 1-SQR_logist/SQT_logist ; R2_logist
R2adj_logist <- 1 -
(((1-R2_logist)*(dim(dados)[1]-1)/(((summary(Modelo_Logist))$df)[2])));
R2adj_logist
# df total=9 df residuo=6
QMe_logist <- SQR_logist/(((summary(Modelo_Logist))$df)[2]); QMe_logist
DPR_logist <- sqrt(QMe_logist) ; DPR_logist
# Gompertz
SQR_gomp <- sum((dados$y-fitted(Modelo_Gompertz))^2) ; SQR_gomp
SQT_gomp <- sum(dados$y*dados$y) - (sum(dados$y)^2)/length(dados$y) ; SQT_gomp
R2_gomp <- 1-SQR_gomp/SQT_gomp ; R2_gomp
R2adj_gomp <- 1 -
(((1-R2_gomp)*(dim(dados)[1]-1)/(((summary(Modelo_Gompertz))$df)[2])));
R2adj_gomp
# df total=9 df residuo=6
QMe_gomp <- SQR_gomp/(((summary(Modelo_Gompertz))$df)[2]); QMe_gomp
DPR_gomp <- sqrt(QMe_gomp) ; DPR_gomp
#########################################################################
#### Arquivo Unico ####
#########################################################################
Resultado <- matrix(NA, nrow = nrow(dados)+1, ncol = 22)
Resultado[1,c(1:6)] <- c("X", "Y", "Predito_log", "Residuos_Log",
"Predito_Gomp", "Residuos_Gomp")
Resultado[c(2:(nrow(dados)+1)),1] <- as.vector(dados$x)
Resultado[c(2:(nrow(dados)+1)),2] <- as.vector(dados$y)
Resultado[c(2:(nrow(dados)+1)),3] <- as.vector(predict(Modelo_Logist))
Resultado[c(2:(nrow(dados)+1)),4] <- as.vector(ResumoModel_Logist$residuals)
Resultado[c(2:(nrow(dados)+1)),5] <- as.vector(predict(Modelo_Gompertz))
Resultado[c(2:(nrow(dados)+1)),6] <- as.vector(ResumoModel_Gompertz$residuals)
Resultado[1, 9] <- "Modelo Logistico"
Resultado[4:6, 9:12] <- ResumoModel_Logist$coefficients
Resultado[4:6, 8] <- rownames(ResumoModel_Logist$coefficients)
Resultado[3, 9:12] <- colnames(ResumoModel_Logist$coefficients)
Resultado[10:12, 9:10] <- IC_logist
Resultado[10:12, 8] <- rownames(IC_logist)
Resultado[9, 9:10] <- colnames(IC_logist)
Resultado[15, 9] <- "Pressupostos"
Resultado[17, 9] <- "Normalidade"
Resultado[18, 9:11] <- c("Teste", "Estatistica", "Valor-p")
Resultado[19, 9:11] <- c(SW_logist$method, SW_logist$statistic,
SW_logist$p.value)
Resultado[20, 9:11] <- c(LL_logist$method, LL_logist$statistic,
LL_logist$p.value)
Resultado[21, 9:11] <- c(KS_logist$method, KS_logist$statistic,
KS_logist$p.value)
Resultado[22, 9:11] <- c(AD_logist$method, AD_logist$statistic,
AD_logist$p.value)
Resultado[25, 9] <- "Homogeneidade"
Resultado[26, 9:11] <- c(BP_logist$method, BP_logist$statistic,
BP_logist$p.value)
Resultado[27, 9:11] <- c(BAR_logist$method, BAR_logist$statistic,
BAR_logist$p.value)
Resultado[29, 9] <- "Independencia"
Resultado[30, 9:11] <- c(BP_logist$method, BP_logist$statistic, BP_logist
$p.value)
Resultado[32, 9] <- "Não-Linearidade"
Resultado[33, 10:11] <- c("Parametrica", "Intrinseca")
Resultado[34, 10:11] <- c(parm_R1_logist, intrinseca_R1_logist)
Resultado[1, 16] <- "Modelo Gompertz"
Resultado[4:6, 16:19] <- ResumoModel_Gompertz$coefficients
Resultado[4:6, 15] <- rownames(ResumoModel_Gompertz$coefficients)
Resultado[3, 16:19] <- colnames(ResumoModel_Gompertz$coefficients)
Resultado[10:12, 16:17] <- IC_gomp
Resultado[10:12, 15] <- rownames(IC_gomp)
Resultado[9, 16:17] <- colnames(IC_gomp)
Resultado[15, 16] <- "Pressupostos"
Resultado[17, 16] <- "Normalidade"
Resultado[18, 16:18] <- c("Teste", "Estatistica", "Valor-p")
Resultado[19, 16:18] <- c(SW_gomp$method, SW_gomp$statistic, SW_gomp$p.value)
Resultado[20, 16:18] <- c(LL_gomp$method, LL_gomp$statistic, LL_gomp$p.value)
Resultado[21, 16:18] <- c(KS_gomp$method, KS_gomp$statistic, KS_gomp$p.value)
Resultado[22, 16:18] <- c(AD_gomp$method, AD_gomp$statistic, AD_gomp$p.value)
Resultado[25, 16] <- "Homogeneidade"
Resultado[26, 16:18] <- c(BP_gomp$method, BP_gomp$statistic, BP_gomp$p.value)
Resultado[27, 16:18] <- c(BAR_gomp$method, BAR_gomp$statistic, BAR_gomp$p.value)
Resultado[29, 16] <- "Independencia"
Resultado[30, 16:18] <- c(BP_gomp$method, BP_gomp$statistic, BP_gomp$p.value)
Resultado[32, 16] <- "Não-Linearidade"
Resultado[33, 17:18] <- c("Parametrica", "Intrinseca")
Resultado[34, 17:18] <- c(parm_R1_gomp, intrinseca_R1_gomp)
Resultado[40, 9:11] <- c("", "Logistico", "Gompertz")
Resultado[41, 9:11] <- c("PIX", PIX_logist, PIX_gomp)
Resultado[42, 9:11] <- c("PIY", PIY_gomp, PIY_gomp)
Resultado[43, 9:11] <- c("PAMX", PAMX_logist, PAMX_gomp)
Resultado[44, 9:11] <- c("PAMY", PAMY_logist, PAMY_gomp)
Resultado[45, 9:11] <- c("PDMX", PDMX_logist, PDMX_gomp)
Resultado[46, 9:11] <- c("PDMY", PDMY_logist, PDMY_gomp)
Resultado[47, 9:11] <- c("PDAX", PDAX_logist, PDAX_gomp)
Resultado[48, 9:11] <- c("PDAY", PDAY_logist, PDAY_gomp)
Resultado[49, 9:11] <- c("SQR", SQR_logist, SQR_gomp)
Resultado[50, 9:11] <- c("SQT", SQT_logist, SQT_gomp)
Resultado[51, 9:11] <- c("QMe", QMe_logist, QMe_gomp)
Resultado[52, 9:11] <- c("DPR", DPR_logist, DPR_gomp)
Resultado[53, 9:11] <- c("R2", R2_logist, R2_gomp)
Resultado[54, 9:11] <- c("R2a", R2adj_logist, R2adj_gomp)
write.table(Resultado, file = "./TesteMFF.txt")
write.csv(Resultado, file = "./TesteMFF.csv")x y
7 0.31
8 0.34
9 0.325
10 0.4175
11 0.505
12 0.405
13 0.36
14 0.6475
15 5
16 0.395
17 0.4825
18 0.5725
19 0.35
20 0.59
21 0.4675
22 0.6525
23 0.3225
24 0.5175
25 0.6675
26 0.4875
27 0.8325
28 0.7925
29 1.0475
30 1.26
31 2.25
32 2.74
33 1.2525
34 1.725
35 2.6575
36 2.7825
37 2.9525
38 2.8
39 4.8
40 3.5175
41 6.0775
42 4.85
43 13.425
44 10.5
45 7.0875
46 9.9425
47 17.1625
48 11.565
49 22.3225
50 40.9075
51 37.48
52 35.455
53 41.22
54 26.3625
55 44.285
56 30.3025
57 12.36
58 48.395
59 20.68
60 43.1325
61 25.6175
62 39.13
63 37.8975
64 40.2
65 31.8975
66 29.5625
67 34.88
68 29.93
69 36.68
70 29.3625
71 24.94
72 31.7125
73 51.8375
74 29.0925
75 41.3525
76 50.125
77 39.7575
78 42.3875
79 47.9625
80 40.61
81 40.365
82 42.0025
83 46.4025
84 37.235
85 18.4875
86 62.4725
87 35.4875
88 31.2175
89 50.7875
90 65.135
91 51.5525
92 49.185
93 49.4825
94 60.775
95 70.125
96 68.835
97 46.685
98 63.3675
99 38.11
100 32.475
x y
7 0.037810738
7 0.013574592
8 0.045045
8 0.011697255
9 0.04247954
9 0.011697255
10 0.050371899
10 0.018773372
11 0.059350651
11 0.024116562
12 0.036584737
12 0.029170931
13 0.038272496
13 0.020795119
14 0.068837337
14 0.037402332
15 0.068041433
15 0.015596339
16 0.05331912
16 0.012708128
17 0.066504556
17 0.014296644
18 0.081250137
18 0.014874287
19 0.04740249
19 0.011119612
20 0.081654446
20 0.017184855
21 0.064909868
21 0.013430181
22 0.090240522
22 0.019062193
23 0.042699145
23 0.011119612
24 0.072814221
24 0.014007823
25 0.093776756
25 0.018195729
26 0.065741754
26 0.01574075
27 0.11989348
27 0.020073066
28 0.113537638
28 0.019639835
29 0.145662985
29 0.029892984
30 0.171490299
30 0.03927967
31 0.306833632
31 0.069605885
32 0.384975832
32 0.074660254
33 0.144967467
33 0.061807715
34 0.231927703
34 0.056320115
35 0.3419044
35 0.100509743
36 0.370725832
36 0.093866858
37 0.362264489
37 0.127370105
38 0.334240761
38 0.129103032
39 0.591844897
39 0.204485339
40 0.393855033
40 0.185423146
41 0.687210221
41 0.314381767
42 0.539884905
42 0.258494885
43 1.198142715
43 0.979969992
44 1.024866673
44 0.688116271
45 0.76044698
45 0.403194256
46 0.990352176
46 0.633817904
47 1.683178821
47 1.117593246
48 1.015772058
48 0.858809541
49 2.16266458
49 1.477319927
50 3.578507895
50 3.050672869
51 3.316700198
51 2.761129716
52 2.89626068
52 2.827269748
53 2.818285293
53 3.776913525
54 2.146582164
54 2.10839403
55 3.984337236
55 3.204036873
56 2.298814349
56 2.573973643
57 0.855332063
57 1.123369668
58 3.561014764
58 4.209278711
59 1.379695141
59 1.925425863
60 3.358142622
60 3.587013651
61 1.683923964
61 2.40761269
62 3.025497512
62 3.272920705
63 2.134729317
63 3.768393302
64 2.017523502
64 4.336215585
65 2.030078693
65 3.057315754
66 1.671818072
66 3.020635474
67 0.415640717
67 4.953570686
68 1.237797797
68 3.464120273
69 1.495946304
69 4.722080574
70 1.365310064
70 3.26367843
71 0.618149138
71 3.255447029
72 1.716752005
72 3.308734522
73 1.231805655
73 6.813722981
74 0.210185246
74 4.25346834
75 2.166259522
75 4.379105518
76 2.530485217
76 5.39301199
77 1.154545392
77 5.0386285
78 1.789288419
78 4.873567241
79 2.502518216
79 5.088016908
80 2.36852312
80 4.085230049
81 2.852617851
81 3.615751351
82 0.571295666
82 5.901914768
83 -0.265324901
83 7.32031519
84 0.350505358
84 5.366007217
85 0.335362245
85 2.522852309
86 1.813143179
86 7.918319278
87 0.857209128
87 4.652185868
88 0.574758053
88 4.252457466
89 1.829650918
89 6.119830288
90 2.928152773
90 7.329557465
91 0.407252363
91 7.506171568
92 3.452788547
92 4.426472179
93 -2.60475807
93 9.878548083
94 2.82462536
94 6.929684652
95 5.116247252
95 6.138314838
96 1.276251755
96 9.368778842
97 2.914381863
97 4.525393405
98 2.283658476
98 7.634985779
99 0.674572796
99 5.215531424
100 -0.461659228
100 5.36947307
_______________________________________________
R-br mailing list
R-br@listas.c3sl.ufpr.br
https://listas.inf.ufpr.br/cgi-bin/mailman/listinfo/r-br
Leia o guia de postagem (http://www.leg.ufpr.br/r-br-guia) e forne�a c�digo
m�nimo reproduz�vel.
