Skip to content
Extraits de code Groupes Projets
Valider f5036761 rédigé par Alain Guillet's avatar Alain Guillet
Parcourir les fichiers

server.R

parent 1770e0ba
Aucune branche associée trouvée
Aucune étiquette associée trouvée
Aucune requête de fusion associée trouvée
Masquer les modifications d'espaces
En ligne Côte à côte
Affichage de
avec 981 ajouts et 0 suppression
server.R 0 → 100644
+ 981
− 0
Voir le fichier @ f5036761
Sys.setlocale("LC_ALL", "fr_FR.UTF-8")#to be sure that accents in text will be allowed in plots
library(shiny)
library(plotrix)
shinyServer(function(input, output){
rv<-reactiveValues() # Create a reactiveValues object, to let us use settable reactive values
rv$last.takesample.value<-0
rv$samples.z<-list()
rv$lastAction<-'none'# To start out, lastAction == NULL, meaning nothing clicked yet
rv$lastDist<-" "
rv$lastp<-0.5
rv$lastminrud<-1
rv$lastmaxrud<-6
rv$lastdf<-5
rv$lastdf1<-5
rv$lastdf2<-20
rv$lastm1<-8
rv$lastm2<-4
rv$lastsd1<-1.5
rv$lastsd2<-1.1
rv$lastN<-0
# An observe block for each button, to record that the action happened
observe({
if (input$takesample != 0) {
rv$lastAction <- 'takesample'
}
})
observe({
if (input$reset != 0) {
rv$lastAction <- 'reset'
rv$last.takesample.value<-0
rv$samples.z<-list()
}
})
getSamples<-reactive({
if(input$takesample > rv$last.takesample.value && rv$lastAction == "takesample"){
return(isolate({#Now do the expensive stuff
samples<-list()
for (i in 1:input$ns){
if (input$dist == "DN") {samples[[i]]<-rnorm(input$n)}
if (input$dist == "DBin") {samples[[i]]<-rbinom(n = 1, size = input$n, prob = input$p)}
#if (input$dist == "DLN"){samples[[i]]<-rlnorm(input$n)}
if (input$dist == "DUD") {x <- min(input$RUD) : max(input$RUD)
samples[[i]]<-sample(x, input$n, replace = TRUE)}
if (input$dist == "DU") {samples[[i]]<-runif (input$n)}
if (input$dist == "DE") {samples[[i]]<-rexp(input$n)}
if (input$dist == "DC") {samples[[i]]<-rchisq(input$n, df = input$df)}
if (input$dist == "DF") {samples[[i]]<-rf(input$n,df1 = input$df1,df2 = input$df2)}
if (input$dist == "DB") {samples[[i]]<-c(rnorm(input$n/2,input$m1, input$sd1), rnorm(input$n/2, input$m2, input$sd2))}
}
return(samples)
}))
} else {
return(NULL)
}})
getPlotHeight <- function() {
if(input$display=="default") {
unit.height<-320 #cannot be auto because height is already "auto" in ui and double auto = conflict
}
if(input$display=="1024") {
unit.height<-280
}
if(input$display=="800") {
unit.height<-250
}
return(2*unit.height)
}
getPlotWidth <- function() {
if(input$display=="default") {
full.plot.width<-1310-400#"auto"
}
if(input$display=="1024") {
full.plot.width<-900-200
}
if(input$display=="800") {
full.plot.width<-700-200
}
if(input$visM && input$display!="default"){
full.plot.width<-full.plot.width+400
}
return(full.plot.width)
}
getInputValues<-reactive({
return(input)#collect all inputs
})
getComputedValues<-reactive({
samples<-list()
samples<-getSamples()
rv$samples.z<-c(rv$samples.z,samples)
v<-getInputValues() # get all values of input list
cv<-list()#created empty computed values list
cv$samples.x<-list()
cv$n.samples<-length(rv$samples.z)
cv$vx <- v$sx^2
## Computation of sample related values ##
if(cv$n.samples>0){
for(i in 1:cv$n.samples){
if (v$dist == "DN") {cv$samples.x[[i]]<-round((rv$samples.z[[i]]*v$sx)+v$mx,2)}
if (v$dist == "DBin") {cv$samples.x[[i]]<-round(rv$samples.z[[i]],2)}
#if (v$dist == "DLN"){cv$samples.x[[i]]<-round((rv$samples.z[[i]]*v$lsx)+v$lmx,2)}
if (v$dist == "DUD") {cv$samples.x[[i]]<-rv$samples.z[[i]]}
if (v$dist == "DU") {cv$samples.x[[i]]<-round(rv$samples.z[[i]]*v$b,2)}
if (v$dist == "DE") {cv$samples.x[[i]]<-round(rv$samples.z[[i]]*(1/v$Lambda),2)}
if (v$dist == "DC") {cv$samples.x[[i]]<-round(rv$samples.z[[i]], 2)}
if (v$dist == "DF") {cv$samples.x[[i]]<-round(rv$samples.z[[i]], 2)}
if (v$dist == "DB") {cv$samples.x[[i]]<-round(rv$samples.z[[i]], 2)}
}
## Automatic reset en cas de modification des paramètres
if (rv$lastDist!=v$dist) {
rv$lastAction <- 'changeDist'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastN!=v$n) {
rv$lastAction <- 'changeN'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastp!=v$p) {
rv$lastAction <- 'changep'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastminrud!=min(v$RUD)) {
rv$lastAction <- 'changeminrud'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastmaxrud!=max(v$RUD)) {
rv$lastAction <- 'changemaxrud'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastdf!=v$df) {
rv$lastAction <- 'changedf'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastdf1!=v$df1) {
rv$lastAction <- 'changedf1'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastdf2!=v$df2) {
rv$lastAction <- 'changedf2'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastm1!=v$m1) {
rv$lastAction <- 'changem1'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastm2!=v$m2) {
rv$lastAction <- 'changem2'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastsd1!=v$sd1) {
rv$lastAction <- 'changesd1'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
if (rv$lastsd2!=v$sd2) {
rv$lastAction <- 'changesd2'
rv$last.takesample.value<-0
rv$samples.z <- list()
cv$samples.x <- list()
}
##Pour passer d'une liste à une matrice
if(v$dist == "DBin"){
cv$samples.x.mat <- matrix(nrow = cv$n.samples, ncol = 1)
cv$samples.p.mat <- matrix(nrow = cv$n.samples, ncol = 1)
for(i in 1:cv$n.samples){
cv$samples.x.mat[i,1] <- cv$samples.x[[i]]
cv$samples.p.mat[i,1] <-cv$samples.x[[i]]/v$n
}
}
else{
cv$samples.x.mat <- matrix(nrow = cv$n.samples, ncol = v$n)
for(i in 1:cv$n.samples){
cv$samples.x.mat[i,] <- cv$samples.x[[i]]
}
}
## Computation of descriptives
cv$samples.x.m.vec<-c() # vector of mean values, each line a sample
cv$samples.x.sd.vec<-c()
cv$samples.x.m.vec<-round(apply(cv$samples.x.mat,1,mean),2)#means of samples
cv$samples.x.sd.vec<-round(apply(cv$samples.x.mat,1,sd),2)#sds of samples
if(v$dist == "DBin"){
cv$samples.p.m.vec<-round(apply(cv$samples.p.mat,1,mean),2)#means of samples
cv$samples.p.sd.vec<-round(apply(cv$samples.p.mat,1,sd),2)#sds of samples
}
## Define subset to plot
cv$samples.x.n.toshow<-0
cv$samples.x.from<-1
if(cv$n.samples>5){
cv$samples.x.from<-cv$n.samples-5+1
}
cv$samples.x.to<-cv$n.samples
if(v$dist =="DBin"){
cv$samples.x.mat.toshow<-matrix(cv$samples.x.mat[cv$samples.x.from:cv$samples.x.to,],ncol=1)
}
else{
cv$samples.x.mat.toshow<-matrix(cv$samples.x.mat[cv$samples.x.from:cv$samples.x.to,],ncol=v$n)
}
cv$samples.x.m.vec.toshow<-cv$samples.x.m.vec[cv$samples.x.from:cv$samples.x.to]
cv$samples.x.sd.vec.toshow<-cv$samples.x.sd.vec[cv$samples.x.from:cv$samples.x.to]
cv$samples.x.i.vec.toshow<-c(cv$samples.x.from:cv$samples.x.to)
cv$samples.x.n.toshow<-length(cv$samples.x.mat.toshow[,1])
cv$samples.x.m.m <- round(mean(cv$samples.x.m.vec),4)
cv$samples.x.v.m <- round(var(cv$samples.x.m.vec),4)
if(v$dist == "DBin"){
cv$samples.p.m.m <- round(mean(cv$samples.p.m.vec),4)
cv$samples.p.v.m <- round(var(cv$samples.p.m.vec),4)
cv$vx<-v$sx^2
cv$lvx<-v$lsx^2
}
## Valeurs qui serviront à définir les limites des axes
hf<-hist(cv$samples.x.mat, freq= TRUE, breaks = 50)
freqcl<- hf$counts
cv$maxfreqcl<-max(freqcl)
n.obs.tot<-length(cv$samples.x.mat)
probcl <- freqcl/n.obs.tot
cv$maxprobcl<-max(probcl)
#if(cv$n.samples <100){breaks =10}
#else{breaks <- sqrt(cv$n.samples)}
#hm<-hist(cv$samples.x.m.vec, freq = TRUE, breaks = breaks)
#cv$freqmcl <- unlist(hm[2])
#densm<-density(cv$samples.x.m.vec)
#cv$highdensm <- unlist(densm[2])
}
## Last takesample value
rv$last.takesample.value<-v$takesample
rv$lastDist<-v$dist
rv$lastp<-v$p
rv$lastminrud<-min(v$RUD)
rv$lastmaxrud<-max(v$RUD)
rv$lastdf<-v$df
rv$lastdf1<-v$df1
rv$lastdf2<-v$df2
rv$lastm1<-v$m1
rv$lastm2<-v$m2
rv$lastsd1<-v$sd1
rv$lastsd2<-v$sd2
rv$lastN<-v$n
return(cv)
})
output$doublePlot <- renderPlot({
v <- getInputValues ()
cv <- getComputedValues ()
par(mfcol = c(2,2))
m<-matrix(c(1,2,3,4),2,2,byrow=TRUE)
layout(m, width=c(4,3)) #
## Set graphic parameters
if(v$display=="default") {
cex.main.title<-2
cex.title<-1.5
cex.samples<-1.5
cex.axis<-1.1
cex.label<-1.2
}
if(v$display=="1024") {
cex.main.title<-1.75
cex.title<-1.2
cex.samples<-1.2
cex.axis<-1
cex.label<-1
}
if(v$display=="800") {
cex.main.title<-1.5
cex.title<-1
cex.samples<-1
cex.axis<-0.8
cex.label<-0.8
}
#Définition des limites de l'axe des abscisses pour le plot
if(v$dist=="DN"&& v$range =="SameRange") {x.lim.inf<-min(v$rangeXdn)
x.lim.sup<-max(v$rangeXdn)}
if(v$dist=="DBin"&& v$range =="SameRange") {x.lim.inf<-min(v$rangeXdbin)
x.lim.sup<-max(v$rangeXdbin)}
#if(v$dist=="DLN"&& v$range =="SameRange"){x.lim.inf<-min(v$rangeXdln)
# x.lim.sup<-max(v$rangeXdln)}
if(v$dist=="DUD"&& v$range =="SameRange") {x.lim.inf<-min(v$rangeXdud)
x.lim.sup<-max(v$rangeXdud)}
if(v$dist=="DU"&& v$range =="SameRange") {x.lim.inf<-min(v$rangeXdu)
x.lim.sup<-max(v$rangeXdu)}
if(v$dist=="DE"&& v$range =="SameRange") {x.lim.inf<-min(v$rangeXde)
x.lim.sup<-max(v$rangeXde)}
if(v$dist=="DC"&& v$range =="SameRange") {x.lim.inf<-min(v$rangeXdc)
x.lim.sup<-max(v$rangeXdc)}
if(v$dist=="DF"&& v$range =="SameRange") {x.lim.inf<-min(v$rangeXdf)
x.lim.sup<-max(v$rangeXdf)}
if(v$dist=="DB"&& v$range =="SameRange") {x.lim.inf<-min(v$rangeXdb)
x.lim.sup<-max(v$rangeXdb)}
if(v$dist=="DN"&& v$range =="DifRange") {Obs.lim.inf<-min(v$rangeObsdn)
Obs.lim.sup<-max(v$rangeObsdn)
Xbar.lim.inf<-min(v$rangeXbardn)
Xbar.lim.sup<-max(v$rangeXbardn)}
if(v$dist=="DBin"&& v$range =="DifRange") {Obs.lim.inf<-min(v$rangeObsdbin)
Obs.lim.sup<-max(v$rangeObsdbin)
Xbar.lim.inf<-min(v$rangeXbardbin)
Xbar.lim.sup<-max(v$rangeXbardbin)}
#if(v$dist=="DLN"&& v$range =="DifRange"){Obs.lim.inf<-min(v$rangeObsdln)
# Obs.lim.sup<-max(v$rangeObsdln)
# Xbar.lim.inf<-min(v$rangeXbardln)
# Xbar.lim.sup<-max(v$rangeXbardln)}
if(v$dist=="DUD"&& v$range =="DifRange") {Obs.lim.inf<-min(v$rangeObsdud)
Obs.lim.sup<-max(v$rangeObsdud)
Xbar.lim.inf<-min(v$rangeXbardud)
Xbar.lim.sup<-max(v$rangeXbardud)}
if(v$dist=="DU"&& v$range =="DifRange") {Obs.lim.inf<-min(v$rangeObsdu)
Obs.lim.sup<-max(v$rangeObsdu)
Xbar.lim.inf<-min(v$rangeXbardu)
Xbar.lim.sup<-max(v$rangeXbardu)}
if(v$dist=="DE"&& v$range =="DifRange") {Obs.lim.inf<-min(v$rangeObsde)
Obs.lim.sup<-max(v$rangeObsde)
Xbar.lim.inf<-min(v$rangeXbarde)
Xbar.lim.sup<-max(v$rangeXbarde)}
if(v$dist=="DC"&& v$range =="DifRange") {Obs.lim.inf<-min(v$rangeObsdc)
Obs.lim.sup<-max(v$rangeObsdc)
Xbar.lim.inf<-min(v$rangeXbardc)
Xbar.lim.sup<-max(v$rangeXbardc)}
if(v$dist=="DF"&& v$range =="DifRange") {Obs.lim.inf<-min(v$rangeObsdf)
Obs.lim.sup<-max(v$rangeObsdf)
Xbar.lim.inf<-min(v$rangeXbardf)
Xbar.lim.sup<-max(v$rangeXbardf)}
if(v$dist=="DB"&& v$range =="DifRange") {Obs.lim.inf<-min(v$rangeObsdb)
Obs.lim.sup<-max(v$rangeObsdb)
Xbar.lim.inf<-min(v$rangeXbardb)
Xbar.lim.sup<-max(v$rangeXbardb)}
#Définition des X conditionnellement à la distribution
if(v$dist=="DN"){X=seq(-10,40, length=1000)}
if(v$dist=="DBin"){X=0:v$n}
#if(v$dist=="DLN"){X=seq(-10,40, length=1000)}
if(v$dist=="DUD"){X= min(v$RUD):max(v$RUD)}
if(v$dist=="DU"){X=seq(-5,25, length=1000)}
if(v$dist=="DE"){X=seq(-5,20, length=1000)}
if(v$dist=="DC"){X=seq(-5,60, length=1000)}
if(v$dist=="DF"){X=seq(-5,10, length=1000)}
#Définition de la densité théorique
getY <-reactive({
if (v$dist=="DN")
return(dnorm(X, mean=v$mx, sd=v$sx))
if (v$dist=="DBin")
return(dbinom(X, size=v$n, prob=v$p))
#if (v$dist=="DLN")
# return(dlnorm(X,meanlog=v$lmx, sdlog =v$lsx))
if (v$dist=="DU")
return(dunif (X, min=0, max=v$b))
if (v$dist=="DE")
return (dexp(X, rate=v$Lambda))
if (v$dist=="DC")
return (dchisq(X, df=v$df))
if (v$dist=="DF")
return(df(X,df1=v$df1, df2=v$df2))
if (v$dist=="DB")
return(density(c(rnorm(1000000/2,v$m1, v$sd1), rnorm(1000000/2, v$m2, v$sd2))))
})
#------------------- Output 1 : ------------------------------
#Afficher les observations pour 5 échantillons prélevés
#Afficher la distribution théorique d'origine (optionnel
#-------------------------------------------------------------
if (v$dist == "DB"){
dens<-getY()
d<-unlist(dens[2])
y.delta<-max(d)
}
else{
if (v$dist == "DUD"){
p <- rep(1/length(X), length(X))
y.delta<-p[1]+p[1]/length(X)
}
else{
y.delta <- max(getY())
}
}
cv$samples.y.mat.toshow<-c()
if(cv$n.samples>0 && cv$samples.x.n.toshow>0){
if(v$dist == "DBin"){
cv$samples.y.mat.toshow<-matrix(rep(y.delta/(5+1)*c(1:cv$samples.x.n.toshow),length(cv$samples.x.mat.toshow[,1])),nrow=length(cv$samples.x.mat.toshow[,1]), ncol = 1)
}
else{
cv$samples.y.mat.toshow<-matrix(rep(y.delta/(5+1)*c(1:cv$samples.x.n.toshow),length(cv$samples.x.mat.toshow[,1])),nrow=length(cv$samples.x.mat.toshow[,1]), ncol = v$n)
}
}
## Définition de pour l'axe des X
if(v$range =="SameRange"){
lim.inf<-x.lim.inf
lim.sup<-x.lim.sup
}
if(v$range =="DifRange"){
lim.inf<-Obs.lim.inf
lim.sup<-Obs.lim.sup
}
range<-lim.sup-lim.inf
## Définition du nb de graduations pour l'axe des X
if(v$dist == "DBin"){
nbgrad<-10
}
else{
if(v$dist == "DUD"){
nbgrad<-range
}
else{
if(range>10){nbgrad <- range}
if(range>5 & range <=10){nbgrad <- range*2}
if(range<=5){nbgrad <- range*4}
}}
## Test about range of 'x'
if(cv$n.samples>0){
if(max(cv$samples.x.mat) > lim.sup || min(cv$samples.x.mat) < lim.inf) {error <-1}
else{error <-0}
}
##############PLOT########################
par(mai=c(0.5,0.8,0.5,0.5))
### CAS N°1 : Si aucune donnée :
if(is.null(cv$samples.x.mat)){
### CAS N°1.1 : Si l'option "afficher la distribution théorique est cochée :
if(v$showreality){
#Si distribution Binomiale
if (v$dist=="DBin"){
Y <- getY()
plot(Y, type = "h",bty="n", xaxs="i",yaxs="i",xlab ="x",cex.lab=cex.label,cex.axis=cex.axis, ylab = expression(P(x)), xlim=c(lim.inf,lim.sup),xaxp=c(lim.inf,lim.sup,nbgrad), lwd=2, col = "red", main = "")
# mtext(bquote(paste("Echantillons prélevés :")), side=3,line=1,adj=0.5, cex=cex.label)
mtext(bquote(paste(X*"~"*Bin(n*","*p)," ",X*"~"*Bin(.(v$n)*","*.(v$p)),sep='')), side=3,line=-1,adj=0.05, cex=cex.label)
}
else{
#Si distribution Uniforme Discrète
if (v$dist=="DUD"){
plot(X, p, col = "red", type = "h",bty="n", xaxs="i",yaxs="i",,cex.lab=cex.label,cex.axis=cex.axis, xlab =" ", ylab=" ",lwd = 2, xlim=c(lim.inf,lim.sup),xaxp=c(lim.inf,lim.sup,nbgrad),ylim = c(0, y.delta), main = "")
# mtext(bquote(paste("Echantillons prélevés :")), side=3,line=1,adj=0.5, cex=cex.label)
mtext(bquote(paste(X*"~"*U*"{"*.(min(v$RUD))*",...,"*.(max(v$RUD))*"}",sep='')), side=3,line=1,adj=-0.1, cex=cex.label)
points(X,p, col = "red", lwd = 2, pch = 19)
lines(X, p, lty = 3)
}
#Pour les autres distributions
else{
plot(c(0),c(-5),lty=1,lwd=1,col="black",yaxt="n",bty="n",las=1,xaxs="i",yaxs="i",cex.lab=cex.label,cex.axis=cex.axis,xlim=c(lim.inf,lim.sup),ylim=c(0,y.delta),xlab="",ylab=" ",xaxp=c(lim.inf,lim.sup,nbgrad),main="")
axis(2,las=2,yaxp=c(0,signif(y.delta,1),5),cex.axis=cex.axis)
if (v$dist == "DB"){
dens <- getY()
lines(dens)
}
else {
Y<-getY()
points(X,Y, type="l")
}
# mtext(bquote(paste("Echantillons prélevés :")), side=3,line=1,adj=0.5, cex=cex.label)
if(v$dist=="DN"){mtext(bquote(paste(X*"~"*N(mu*","*sigma^2) ," ", X*"~"*N(.(v$mx)*","*.(cv$vx)),sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
if(v$dist=="DU"){mtext(bquote(paste(X*"~"*U(theta[1]*","*theta[2]) ," ", X*"~"*U(.0*","*.(v$b)),sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
if(v$dist=="DE"){mtext(bquote(paste(X*"~"*E(lambda) ," ", X*"~"*E(.(v$Lambda)),sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
if(v$dist=="DC"){mtext(bquote(paste(X*"~"*chi^2, (nu)," ", X*"~"*chi^2,(.(v$df)),sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
if(v$dist=="DF"){mtext(bquote(paste(X*"~"*F[nu[1]*","*nu[2]] ," ", X*"~"*F[.(v$df1)*","*.(v$df2)],sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
}}
}
###CAS N°1.2 : Si l'option "afficher la distribution théorique" n'est pas cochée :
else{
#par(mai=c(0.5,0.8,0.5,0.5))
plot(c(0),c(-5),lty=1,lwd=1,col="black",yaxt="n",bty="n",las=1,xaxs="i",yaxs="i",cex.lab=cex.label,cex.axis=cex.axis,xlim=c(lim.inf,lim.sup),ylim=c(0,y.delta),xlab="",ylab=" ",xaxp=c(lim.inf,lim.sup,nbgrad),main="")
# mtext(bquote(paste("Echantillons prélevés :")), side=3,line=1,adj=0.5, cex=cex.label)
}
}
##CAS N°2 : Si 1 ou plusieurs échantillons ont déjà été tirés :
else{
### CAS N°2.1 : Si conflit entre limites des X et observations prélevées : afficher un msg d'erreur
if(error==1){
plot(1:10,1:10, col = "white", xlab="",ylab="",xaxt="n",yaxt="n",bty="n",type='l')
text(5,8, labels = bquote("Certaines valeurs dépassent les limites défines en abscisse."), cex = cex.label, col = "red")
text(5,7, labels = bquote("Modifiez le choix de l'étendue au moyen du slider adéquat."), cex = cex.label, col = "red")
}
### CAS N°2.2 : Si pas d'erreur :
if(error==0){
### CAS N°2.2.1 : Si l'option "afficher la distribution théorique" est cochée :
if(v$showreality){
##Si la distribution est binomiale :
if (v$dist=="DBin"){
Y <- getY()
plot(Y, type = "h",bty="n", xaxs="i",yaxs="i",xlab ="x", ylab = expression(P(x)),cex.lab=cex.label,cex.axis=cex.axis, xlim=c(lim.inf,lim.sup),xaxp=c(lim.inf,lim.sup,nbgrad), lwd=2, col = "red", main = "")
# mtext(bquote(paste("Echantillons prélevés :")), side=3,line=1,adj=0.5, cex=cex.label)
mtext(bquote(paste(X*"~"*Bin(n*","*p)," ",X*"~"*Bin(.(v$n)*","*.(v$p)),sep='')), side=3,line=-1,adj=0.05, cex=cex.label)
for(i in 1:cv$samples.x.n.toshow){
points(cv$samples.x.mat.toshow[i,],cv$samples.y.mat.toshow[i,],cex=cex.samples*0.8)
text(cv$samples.x.m.vec.toshow[i],cv$samples.y.mat.toshow[i,1],labels=bquote(x[.(cv$samples.x.i.vec.toshow[i])]),cex=cex.samples*1.2,col="blue")
}
}
else{
##Si la distribution est uniforme discète :
if (v$dist=="DUD"){
plot(X, p, col = "red", type = "h",bty="n", xaxs="i",yaxs="i", ylab=" ",cex.lab=cex.label,cex.axis=cex.axis,lwd = 2, xlim=c(lim.inf,lim.sup),xaxp=c(lim.inf,lim.sup,nbgrad),ylim = c(0, y.delta), main = "")
mtext(bquote(paste("Distribution théorique")), side=3,line=1,adj=0.5, cex=cex.label)
mtext(bquote(paste(X*"~"*U*"{"*.(min(v$RUD))*",...,"*.(max(v$RUD))*"}",sep='')), side=3,line=1,adj=-0.1, cex=cex.label)
points(X,p, col = "red", lwd = 2, pch = 19)
lines(X, p, lty = 3)
for(i in 1:cv$samples.x.n.toshow){
points(jitter(cv$samples.x.mat.toshow[i,],0.5),jitter(cv$samples.y.mat.toshow[i,],0.5),cex=cex.samples*0.8)
text(cv$samples.x.m.vec.toshow[i],cv$samples.y.mat.toshow[i,1],labels=bquote(bar(x)[.(cv$samples.x.i.vec.toshow[i])]),cex=cex.samples*1.2,col="blue")
}
}
##Pour toutes les autres distributions :
else{
plot(c(0),c(-5),lty=1,lwd=1,col="black",yaxt="n",bty="n",las=1,xaxs="i",yaxs="i",cex.lab=cex.label,cex.axis=cex.axis,xlim=c(lim.inf,lim.sup),ylim=c(0,y.delta),xlab="",ylab=" ",xaxp=c(lim.inf,lim.sup,nbgrad),main="")
# mtext(bquote(paste("Echantillons prélevés :")), side=3,line=1,adj=0.5, cex=cex.label)
for(i in 1:cv$samples.x.n.toshow){
points(cv$samples.x.mat.toshow[i,],cv$samples.y.mat.toshow[i,],cex=cex.samples*0.8)
text(cv$samples.x.m.vec.toshow[i],cv$samples.y.mat.toshow[i,1],labels=bquote(bar(x)[.(cv$samples.x.i.vec.toshow[i])]),cex=cex.samples*1.2,col="blue")
}
axis(2,las=2,yaxp=c(0,signif(y.delta,1),5),cex.axis=cex.axis)
if (v$dist == "DB"){
dens <- getY()
lines(dens)
}
else {
Y<-getY()
points(X,Y, type="l")
}
if(v$dist=="DN"){mtext(bquote(paste(X*"~"*N(mu*","*sigma^2) ," ", X*"~"*N(.(v$mx)*","*.(v$sx)),sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
if(v$dist=="DU"){mtext(bquote(paste(X*"~"*U(theta[1]*","*theta[2]) ," ", X*"~"*U(.0*","*.(v$b)),sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
if(v$dist=="DE"){mtext(bquote(paste(X*"~"*E(lambda) ," ", X*"~"*E(.(v$Lambda)),sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
if(v$dist=="DC"){mtext(bquote(paste(X*"~"*chi^2, (nu)," ", X*"~"*chi^2,(.(v$df)),sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
if(v$dist=="DF"){mtext(bquote(paste(X*"~"*F[nu[1]*","*nu[2]] ," ", X*"~"*F[.(v$df1)*","*.(v$df2)],sep='')), side=3,line=1,adj=-0.1, cex=cex.label)}
}
}
}
###CAS N°2.2.2 : Si la case "Afficher la distribution théorique n'est pas cochée":
else{
##Si la distribution est binomiale :
if (v$dist=="DBin"){
plot(c(0),c(-5),lty=1,lwd=1,col="black",yaxt="n",bty="n",las=1,xaxs="i",yaxs="i",cex.lab=cex.label,cex.axis=cex.axis,xlim=c(lim.inf,lim.sup),ylim=c(0,y.delta),xlab="",ylab=" ",xaxp=c(lim.inf,lim.sup,nbgrad), main="") #,
# mtext(bquote(paste("Echantillons prélevés :")), side=3,line=1,adj=0.5, cex=cex.label)
for(i in 1:cv$samples.x.n.toshow){
points(cv$samples.x.mat.toshow[i,],cv$samples.y.mat.toshow[i,],cex=cex.samples*0.8)
text(cv$samples.x.m.vec.toshow[i],cv$samples.y.mat.toshow[i,1],labels=bquote(x[.(cv$samples.x.i.vec.toshow[i])]),cex=cex.samples*1.2,col="blue")
}
}
else{
plot(c(0),c(-5),lty=1,lwd=1,col="black",yaxt="n",bty="n",las=1,xaxs="i",yaxs="i",cex.lab=cex.label,cex.axis=cex.axis,xlim=c(lim.inf,lim.sup),ylim=c(0,y.delta),xlab="",ylab=" ",xaxp=c(lim.inf,lim.sup,nbgrad), main="")
# mtext(bquote(paste("Echantillons prélevés :")), side=3,line=1,adj=0.5, cex=cex.label)
for(i in 1:cv$samples.x.n.toshow){
if(v$dist=="DUD"){points(jitter(cv$samples.x.mat.toshow[i,],0.5),jitter(cv$samples.y.mat.toshow[i,],0.5),cex=cex.samples*0.8)}
else{points(cv$samples.x.mat.toshow[i,],cv$samples.y.mat.toshow[i,],cex=cex.samples*0.8)}
text(cv$samples.x.m.vec.toshow[i],cv$samples.y.mat.toshow[i,1],labels=bquote(bar(x)[.(cv$samples.x.i.vec.toshow[i])]),cex=cex.samples*1.2,col="blue")
}
}}}
}
#------------------- Output 2 : --------------------------------------
#Afficher les stats descriptives des échantillons prélevés (optionnel)
#---------------------------------------------------------------------
### CAS n°1 : Si aucune donnée :
if(is.null(cv$samples.x.mat)){
par(mai=c(0.5,0,0.5,0))
plot(c(0,1),c(0,0),col="white",xaxt="n",yaxt="n",xlab="",ylab="",ylim=c(0,y.delta),bty="n",las=1)
mtext(bquote(paste("Descriptives : ", N == .(0), sep="")),side=3,line=1,adj=0, at=0.00, cex=cex.label)
}
### CAS n°2 : Si 1 ou plusieurs échantillons ont déjà été tirés :
else{
par(mai=c(0.5,0,0.5,0))
plot(c(0,1),c(0,0),col="white",xaxt="n",yaxt="n",xlab="",ylab="",ylim=c(0,y.delta),bty="n",las=1)
### CAS N°2.1 : Si l'option "afficher les stat descr" est cochée :
if(v$empPl){
mtext(bquote(paste("Descriptives : ", N == .(cv$n.samples), sep="")),side=3,line=1,adj=0, at=0.00, cex=cex.label)
if(cv$samples.x.n.toshow>0){
if(v$dist =="DBin"){
for(i in 1:cv$samples.x.n.toshow){
text(0,cv$samples.y.mat.toshow[i,1],labels=bquote(paste(x[.(cv$samples.x.i.vec.toshow[i])] == .(sprintf("%.2f",cv$samples.x.mat.toshow[i])),sep="")),col="blue",pos=4, cex=cex.samples)
text(0.3,cv$samples.y.mat.toshow[i,1],labels=bquote(paste(p[.(cv$samples.x.i.vec.toshow[i])] == .(sprintf("%.2f",cv$samples.x.mat.toshow[i]/v$n)),sep="")),pos=4,cex=cex.samples)
}
}
else{
for(i in 1:cv$samples.x.n.toshow){
text(0,cv$samples.y.mat.toshow[i,1],labels=bquote(paste(bar(x)[.(cv$samples.x.i.vec.toshow[i])] == .(sprintf("%.2f",cv$samples.x.m.vec.toshow[i])),sep="")),col="blue",pos=4, cex=cex.samples)
text(0.3,cv$samples.y.mat.toshow[i,1],labels=bquote(paste(s[.(cv$samples.x.i.vec.toshow[i])] == .(sprintf("%.2f",cv$samples.x.sd.vec.toshow[i])),sep="")),pos=4,cex=cex.samples)
}
}
if(cv$n.samples>1 && v$dist!="DBin"){
mtext(bquote(paste("E(",bar(X),")" == .(cv$samples.x.m.m), sep="")),side=1,line=1,adj=0, at=0.01, cex=cex.label)
mtext(bquote(paste("V(",bar(X),")" == .(cv$samples.x.v.m), sep="")),side=1,line=1,adj=0, at=0.31, cex=cex.label)
}
if(cv$n.samples>1 && v$dist=="DBin"){
mtext(bquote(paste("E(",X,")" == .(cv$samples.x.m.m), sep="")),side=1,line=-1,adj=0, at=0.01, cex=cex.label)
mtext(bquote(paste("V(",X,")" == .(cv$samples.x.v.m), sep="")),side=1,line= 1,adj=0, at=0.01, cex=cex.label)
mtext(bquote(paste("E(",p,")" == .(cv$samples.p.m.m), sep="")),side=1,line=-1,adj=0, at=0.31, cex=cex.label)
mtext(bquote(paste("V(",p,")" == .(cv$samples.p.v.m), sep="")),side=1,line= 1,adj=0, at=0.31, cex=cex.label)
}
}}
### CAS N°2.2 : Si l'option "afficher les stat descr" n'est pas cochée :
else {}
}
#------------------- Output 3 : --------------------------------------
#Histogramme des données d'échantillonnage
#Afficher leur distribution (optionnel)
#---------------------------------------------------------------------
##Définition des limites pour l'axe des X
if(v$range =="SameRange"){
lim.inf<-x.lim.inf
lim.sup<-x.lim.sup
}
if(v$range =="DifRange"){
lim.inf<-Obs.lim.inf
lim.sup<-Obs.lim.sup
}
range<-lim.sup-lim.inf
## Définition du nb de graduations pour l'axe des X
if(v$dist == "DBin"){
nbgrad<-10
}
else{
if(v$dist == "DUD"){
nbgrad<-range
}
else{
if(range>10){nbgrad <- range}
if(range>5 & range <=10){nbgrad <- range*2}
if(range<=5){nbgrad <- range*4}
}}
## Test about range of 'x'
if(cv$n.samples>0){
if(max(cv$samples.x.mat) > lim.sup || min(cv$samples.x.mat) < lim.inf) {error <-1}
else{error <-0}
}
##############PLOT########################
par(mai=c(0.5,0.8,0.5,0.5))
### CAS N°1 : Si aucune donnée :
if(is.null(cv$samples.x.mat)){
Y <- c()
X <-c()
plot(X, Y, main="",yaxt="n",bty="n",cex.axis=cex.axis,cex.lab = cex.label,xlim=c(lim.inf,lim.sup),ylim=c(0,y.delta),xlab="", ylab = "",xaxp=c(lim.inf,lim.sup,nbgrad))
if(v$dist == "DBin"){mtext(bquote(paste("Distribution du nombre de succès (N tentatives)")), side=3,line=1,adj=0.5, cex=cex.label)}
else{
if(v$dist == "DUD"){mtext(bquote(paste("Distribution des données d'échantillonnage")), side=3,line=1,adj=0.5, cex=cex.label)}
else{mtext(bquote(paste("Histogramme des données d'échantillonnage")), side=3,line=1,adj=0.5, cex=cex.label)}
}
}
##CAS N°2 : Si 1 ou plusieurs échantillons ont déjà été tirés :
else{
### CAS N°2.1 : Si conflit entre limites des X et observations prélevées : afficher un msg d'erreur
if(error==1){
plot(1:10,1:10, col = "white", xlab="",ylab="",xaxt="n",yaxt="n",bty="n",type='l')
text(5,8, labels = bquote("Certaines valeurs dépassent les limites défines en abscisse."), cex = cex.label, col = "red")
text(5,7, labels = bquote("Modifiez le choix de l'étendue au moyen du slider adéquat."), cex = cex.label, col = "red")
}
### CAS N°2.2 : Si pas d'erreur
if(error==0){
#Si la distribution est Binomiale & que l'option "afficher la densité normale" est cochée :
if (v$dist=="DBin"&&v$showNdensity) {
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
hist(cv$samples.x.mat, probability=TRUE,yaxt="n",bty="n",xaxs="i",yaxs="i",xlab="",ylab=HTML("Densité"), xlim=c(lim.inf,lim.sup),xaxp=c(lim.inf,lim.sup,nbgrad), col = 'grey',main = "", breaks = 50, cex.lab = cex.label) #, ylim=c(0,cv$maxfreqcl*1.1)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Distribution du nombre de succès (N tentatives)")), side=3,line=1, adj=0.5, cex=cex.label)
if(cv$n.samples>1){
mtext(bquote(paste(X%~~%N(np*","*np(1-p)),sep='')), side=3,line=-1,adj=0.05, cex=cex.label)
mtext(bquote(paste(X%~~%N(.(cv$samples.x.m.m)*","*.(cv$samples.x.v.m)),sep='')), side=3,line=-3,adj=0.05, cex=cex.label)
lim_dens_inf <- min (cv$samples.x.mat)-1
lim_dens_sup <- max(cv$samples.x.mat)+1
xfit<-seq(lim_dens_inf,lim_dens_sup,length=1000)
yfit<-dnorm(xfit,mean=mean(cv$samples.x.mat),sd=sd(cv$samples.x.mat))
lines(xfit, yfit, col="blue", type = 'l',lwd=2)
}}
#Si la distribution est Binomiale mais que l'option "afficher la densité normale" n'est pas cochée :
else {
if(v$dist=="DBin"){
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
hist(cv$samples.x.mat, freq=TRUE,yaxt="n",bty="n",xaxs="i",yaxs="i",xlab="",ylab=HTML("Fréquences"), xlim=c(lim.inf,lim.sup), ylim=c(0,cv$maxfreqcl*1.1),xaxp=c(lim.inf,lim.sup,nbgrad), col = 'grey',main = "", breaks = 50, cex.lab = cex.label)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Distribution du nombre de succès (N tentatives)")), side=3,line=1, adj=0.5, cex=cex.label)
}
#Si la distribution est Uniforme discrète et que l'option "afficher la distribution théorique" est cochée:
else{
if (v$dist=="DUD"){
if(v$showreality){
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
b<-barplot(prop.table(table(cv$samples.x.mat)), bty="n", yaxt="n", col = 'grey',main = "",space = 2,xlab="", ylab=HTML("Fréquences relatives"), cex.lab = cex.label)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Distribution des données d'échantillonnage")), side=3,line=1, adj=0.5, cex=cex.label)
abline(h=p, lty = 3, lwd = 2)
}
#Si la distribution est Uniforme discrète et que l'option "afficher la distribution théorique" n'est pas cochée:
else{
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
b<-barplot(table(cv$samples.x.mat), bty="n", yaxt="n", col = 'grey',main = "",space = 2,xlab="", ylab=HTML("Fréquences"), cex.lab = cex.label)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Distribution des données d'échantillonnage")), side=3,line=1, adj=0.5, cex=cex.label)
}
}
#Pour toutes les autres distributions quand l'option "afficher la distribution théorique" est cochée:
else{
if(v$showreality){
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
hist(cv$samples.x.mat, probability=TRUE,yaxt="n",bty="n", xaxs="i",yaxs="i",xlab="", ylab=HTML("Densité"),xlim=c(lim.inf,lim.sup),xaxp=c(lim.inf,lim.sup,nbgrad),col = 'grey',main = "",breaks=50, cex.lab=cex.label) #,ylim =c(0, max(c(y.delta, cv$maxprobcl))*1.1)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Histogramme des données d'échantillonnage")), side=3,line=1,adj=0.5, cex=cex.label)
#afficher la distribution théorique
if (v$dist == "DB"){lines(getY())}
else{lines(X, getY(), type = 'l')}
}
#Pour toutes les autres distributions quand l'option "afficher la distribution théorique" n'est pas cochée:
else{
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
hist(cv$samples.x.mat, freq=TRUE,yaxt="n",bty="n",xaxs="i",yaxs="i",xlab="",ylab=HTML("Fréquences"), xlim=c(lim.inf,lim.sup), ylim=c(0,cv$maxfreqcl*1.1),xaxp=c(lim.inf,lim.sup,nbgrad), col = 'grey',main = "", breaks = 50, cex.lab = cex.label)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Histogramme des données d'échantillonnage")), side=3,line=1,adj=0.5, cex=cex.label)
}
}}
}
}}
#------------------- Output 4 : --------------------------------------
#Histogramme des moyennes d'échantillonnage
#Afficher leur distribution (optionnel)
#---------------------------------------------------------------------
##Définition des limites pour l'axe des X
if(v$range =="SameRange"){
lim.inf<-x.lim.inf
lim.sup<-x.lim.sup
}
if(v$range =="DifRange"){
lim.inf<-Xbar.lim.inf
lim.sup<-Xbar.lim.sup
}
range <-lim.sup-lim.inf
## Définition du nb de graduations pour l'axe des X
if(v$dist == "DBin"){
nbgrad<-10
}
else{
if(v$dist == "DUD"){
nbgrad<-range
}
else{
if(range>10){nbgrad <- range}
if(range>5 & range <=10){nbgrad <- range*2}
if(range<=5){nbgrad <- range*4}
}
}
##Définition du nb d'intervalles pour l'histogramme
if(v$dist =="DE" || v$dist =="DF") {
breaks<-seq(lim.inf, lim.sup, 0.01)
}
else {
breaks<-seq(lim.inf, lim.sup, 0.05)
}
## Test about range of 'x'
if(v$dist == "DBin"){
if(cv$n.samples>0){
if(max(cv$samples.p.mat) > lim.sup || min(cv$samples.p.mat) < lim.inf) {error <-1}
else{error <-0}
}
}
else{
if(cv$n.samples>0){
#for (i in 1: length(cv$samples.x.m.vec)){
if(max(cv$samples.x.m.vec) > lim.sup || min(cv$samples.x.m.vec)< lim.inf) {error <-1}
else{error <-0}
#}
}
}
##############PLOT########################
par(mai=c(0.5,0.8,0.5,0.5))
### CAS N°1 : Si aucune donnée :
if(is.null(cv$samples.x.mat)){
Y <- c()
X <-c()
par(mai=c(0.5,0.8,0.5,0), xaxs="i",yaxs="i")
plot(X, Y, main="",yaxt="n",bty="n",xlab="", ylab = "", xlim=c(lim.inf,lim.sup),ylim=c(0,y.delta), cex.axis=cex.axis,cex.lab = cex.label, ,xaxp=c(lim.inf,lim.sup,nbgrad))
if(v$dist == "DBin"){mtext(bquote(paste("Distribution de la proportion de succès")), side=3,line=1,adj=0.5, cex=cex.label)}
else{mtext(bquote(paste("Histogramme des moyennes d'échantillonnage")), side=3,line=1,adj=0.5, cex=cex.label)}
}
##CAS N°2 : Si 1 ou plusieurs échantillons ont déjà été tirés :
else{
### CAS N°2.1 : Si conflit entre limites des X et observations prélevées : afficher un msg d'erreur
if(error==1){
plot(1:10,1:10, col = "white", xlab="",ylab="",xaxt="n",yaxt="n",bty="n",type='l')
text(5,8, labels = bquote("Certaines valeurs dépassent les limites défines en abscisse."), cex = cex.label*3/4, col = "red")
text(5,7, labels = bquote("Modifiez le choix de l'étendue au moyen du slider adéquat."), cex = cex.label*3/4, col = "red")
}
### CAS N°2.2 : Si pas d'erreur
if(error==0){
#Si la distribution est Binomiale & que l'option "afficher la densité normale" est cochée :
if (v$dist=="DBin"){
if(v$showNdensity){
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
hist(cv$samples.p.mat, probability=TRUE,yaxt="n",bty="n",xaxs="i",yaxs="i",xlab="",ylab=HTML("Densité"), xlim=c(lim.inf,lim.sup),xaxp=c(lim.inf,lim.sup,nbgrad), col = 'grey',main = "", breaks = 50, cex.lab = cex.label) #, ylim=c(0,cv$maxfreqcl*1.1)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Distribution de la proportion de succès")), side=3,line=1, adj=0.5, cex=cex.label)
if(cv$n.samples>1){
lim_dens_inf <- min (cv$samples.p.mat)-0.1
lim_dens_sup <- max(cv$samples.p.mat)+0.1
xfit<-seq(lim_dens_inf,lim_dens_sup,length=1000)
yfit<-dnorm(xfit,mean=mean(cv$samples.p.mat),sd=sd(cv$samples.p.mat))
lines(xfit, yfit, col="blue", type = 'l',lwd=2)
mtext(bquote(paste(bar(X)%~~%N(p*","*p(1-p)/n),sep='')), side=3,line=-1,adj=0.05, cex=cex.label)
mtext(bquote(paste(bar(X)%~~%N(.(cv$samples.p.m.m)*","*.(cv$samples.p.v.m)),sep='')), side=3,line=-3,adj=0.05, cex=cex.label)
}
}
#Si la distribution est Binomiale mais que l'option "afficher la densité normale" n'est pas cochée :
else {
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
hist(cv$samples.x.mat/v$n, freq=TRUE,yaxt="n",bty="n",xaxs="i",yaxs="i",xlab="",ylab=HTML("Fréquences"), xlim=c(lim.inf,lim.sup), ylim=c(0,cv$maxfreqcl*1.1),xaxp=c(lim.inf,lim.sup,nbgrad), col = 'grey',main = "", breaks = 50, cex.lab = cex.label)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Distribution de la proportion de succès")), side=3,line=1, adj=0.5, cex=cex.label)
}
}
#Pour toutes les autres distributions quand l'option "afficher la densité normale sur l'histogramme des moyennes" est cochée:
else{
if(v$showMdensity){
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
hist(cv$samples.x.m.vec, probability=TRUE,yaxt="n",bty="n", xaxs="i",yaxs="i",xlab="", ylab=HTML("Densité"),xlim=c(lim.inf,lim.sup),xaxp=c(lim.inf,lim.sup,nbgrad),col = 'grey',main = "",breaks=breaks, cex.lab=cex.label)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Histogramme des moyennes d'échantillonnage")), side=3,line=1,adj=0.5, cex=cex.label)
if(cv$n.samples>1){
mtext(bquote(paste(bar(X)%~~%N(E(bar(X))*","*V(bar(X))),sep='')), side=3,line=-1,adj=0.05, cex=cex.label)
mtext(bquote(paste(bar(X)%~~%N(.(cv$samples.x.m.m )*","*.(cv$samples.x.v.m )),sep='')), side=3,line=-3,adj=0.05, cex=cex.label)
lim_inf <- min (cv$samples.x.m.vec)-1
lim_sup <- max(cv$samples.x.m.vec)+1
xfit<-seq(lim_inf,lim_sup,length=1000)
yfit<-dnorm(xfit,mean=mean(cv$samples.x.m.vec),sd=sd(cv$samples.x.m.vec))
lines(xfit, yfit, col="blue", type = 'l',lwd=2)
}
}
#Pour toutes les autres distributions quand l'option "afficher la densité normale sur l'histogramme des moyennes" n'est pas cochée:
else{
par(mai=c(0.5,0.8,0.5,0.5), xaxs="i",yaxs="i")
h<-hist(cv$samples.x.m.vec, freq = TRUE, yaxt="n",bty="n", xaxs="i",yaxs="i", xlab="", ylab=HTML("Fréquences"), xlim=c(lim.inf, lim.sup), xaxp=c(lim.inf,lim.sup,nbgrad), col='grey', main="", breaks=breaks, cex.lab = cex.label)
axis(2,las=2,cex.axis=cex.axis)
mtext(bquote(paste("Histogramme des moyennes d'échantillonnage")), side=3,line=1,adj=0.5, cex=cex.label)
}
}
}}
},height = getPlotHeight, width=getPlotWidth)
})
0% ou .
You are about to add 0 people to the discussion. Proceed with caution.
Terminez d'abord l'édition de ce message.
Veuillez vous inscrire ou vous pour commenter