Objectifs

• Calculer pour chaque mois de l’année la moyenne et son intervalle de confiance sur la température entre 8h et 9h du matin
• Nous utiliserons les données de bdd.csv
• Faire une représentation graphique
• Ajouter les quartiles à 25 et 75%

Etape 01 : charger les données

bdd <- read.table("./DATA/bdd.csv", header = TRUE, sep = ",", dec = ".")
head(bdd)

##   year month day hour minute second temperature      gas humidity pressure
## 1 2018    10  22   17     58      1    10.75793 11811196 80.34636 1040.177
## 2 2018    10  22   17     59     26    10.76887 11682953 79.98910 1040.153
## 3 2018    10  22   18      0     29    10.80402 11569892 79.74389 1040.090
## 4 2018    10  22   18      1     31    10.78391 11582347 79.84094 1040.164
## 5 2018    10  22   18      2     33    10.73586 11582347 80.57278 1040.286
## 6 2018    10  22   18      3     36    10.66633 11582347 81.02920 1040.477
##   lightVisible lightIR lightUV
## 1          264     318       4
## 2          266     313       5
## 3          263     306       3
## 4          267     307       4
## 5          266     308       5
## 6          266     307       5

Etape 02 : faire une liste avec pour chaque mois les températures entre 8 et 9h

bdd8list <- split(
  bdd$temperature[bdd$hour == 8], 
  f = bdd$month[bdd$hour == 8])

Etape 02 : faire une liste avec pour chaque mois les températures entre 8 et 9h

str(bdd8list)

## List of 12
##  $ 1 : num [1:1791] 5.21 5.21 5.23 5.22 5.22 ...
##  $ 2 : num [1:1616] 0.49 0.486 0.455 0.407 0.353 ...
##  $ 3 : num [1:1495] 5.33 5.28 5.28 5.3 5.31 ...
##  $ 4 : num [1:1523] 3.95 3.97 4 4 4.01 ...
##  $ 5 : num [1:1781] 4.31 4.43 4.53 4.63 4.72 ...
##  $ 6 : num [1:1708] 16.5 16.6 16.6 16.5 16.5 ...
##  $ 7 : num [1:1769] 15.1 15 15.1 15.2 15.3 ...
##  $ 8 : num [1:1767] 14.2 14.2 14.3 14.4 14.5 ...
##  $ 9 : num [1:1702] 14.1 14 13.9 13.9 13.9 ...
##  $ 10: num [1:688] 2.49 2.47 2.43 2.44 2.46 ...
##  $ 11: num [1:1735] 4.8 4.78 4.8 4.82 4.86 ...
##  $ 12: num [1:1790] 4.13 4.07 4 4.02 4.01 ...

Etape 03 : calculer la moyenne + IC95%

resCI95 <- sapply(seq_along(bdd8list), function(i){
  ci <- t.test(bdd8list[[i]], conf.level = 0.95)$conf.int
  moy <- mean(bdd8list[[i]])
  return(c(moy, ci))
})
colnames(resCI95) <- names(bdd8list)
rownames(resCI95) <- c("moy", "CI-", "CI+")

Etape 03 : calculer la moyenne + IC95%

print(resCI95)

##             1         2        3        4        5        6        7
## moy 0.5091296 0.7647209 7.421766 8.188332 11.19676 17.16307 18.49170
## CI- 0.3538382 0.6214972 7.198553 7.958890 11.02802 16.93950 18.34085
## CI+ 0.6644210 0.9079446 7.644980 8.417774 11.36549 17.38664 18.64254
##            8        9       10       11       12
## moy 15.33637 11.90063 5.383014 3.891985 2.825919
## CI- 15.20963 11.76318 5.044672 3.709193 2.607323
## CI+ 15.46312 12.03808 5.721356 4.074776 3.044515

Etape 04 : représentation graphique

plot(x = 1:12, y = resCI95[1, ], ylim = c(0, 25),
  xlab = "Month", ylab = "Temperature from 8:00 to 8:59 (°C)",
  type = "l", lwd = 2, panel.first = {
    polygon(
      x = c(1:12, 12:1), 
      y = c(resCI95[2, ], rev(resCI95[3, ])), 
      col = 2, border = NA)
  })

Etape 04 : rep. graph.

Etape 04 : rep. graph.

par(mar = c(4, 4, 1, 1))
plot(x = 1:12, y = resCI95[1, ], ylim = c(0, 25),
  xlab = "Month", ylab = "Temperature from 8:00 to 8:59 (°C)",
  type = "l", lwd = 2, axes = FALSE, panel.first = {
    grid()
    polygon(
      x = c(1:12, 12:1), 
      y = c(resCI95[2, ], rev(resCI95[3, ])), 
      col = 2, border = NA)
  })
axis(1)
axis(2)

Etape 04 : rep. graph.

Etape 04 : rep. graph.

par(mar = c(4, 4, 1, 1))
plot(x = 1:12, y = resCI95[1, ], ylim = c(0, 25),
  xlab = "Month", ylab = "Temperature from 8:00 to 8:59 (°C)",
  type = "l", lwd = 2, axes = FALSE, panel.first = {
    grid()
    polygon(
      x = c(1:12, 12:1), 
      y = c(resCI95[2, ], rev(resCI95[3, ])), 
      col = 2, border = NA)
  })
axis(1, at = 1:12, labels = month.abb, las = 3)
axis(2)

Etape 04 : rep. graph.

Etape 04 : rep. graph.

par(mar = c(4, 4, 1, 1))
plot(x = 1:12, y = resCI95[1, ], ylim = c(0, 25),
  xlab = "Month", ylab = "Temperature from 8:00 to 8:59 (°C)",
  type = "l", lwd = 2, axes = FALSE, panel.first = {
    grid()
    polygon(
      x = c(1:12, 12:1), 
      y = c(resCI95[2, ], rev(resCI95[3, ])), 
      col = 2, border = NA)
  })
axis(1, at = 1:12, labels = month.abb, las = 3)
axis(2)
text(x = 1:12, y = resCI95[1, ], 
  labels = round(resCI95[1, ]), 
  pos = 1)

Etape 04 : rep. graph.

Etape 04 : rep. graph.

par(mar = c(4, 4, 1, 1))
plot(x = 1:12, y = resCI95[1, ], ylim = c(0, 25),
  xlab = "Month", ylab = "Temperature from 8:00 to 8:59 (°C)",
  type = "l", lwd = 2, axes = FALSE, panel.first = {
    grid()
    polygon(
      x = c(1:12, 12:1), 
      y = c(resCI95[2, ], rev(resCI95[3, ])), 
      col = 2, border = NA)
  })
axis(1, at = 1:12, labels = month.abb, las = 3)
axis(2)
text(x = 1:12, y = resCI95[1, ], 
  labels = round(resCI95[1, ]), 
  pos = c(1, 1, 2, 4, 4, 2, 3, 1, 4, 1, 1, 1))

Etape 04 : rep. graph.

Etape 05 : quartiles

resCI95 <- sapply(seq_along(bdd8list), function(i){
  ci <- t.test(bdd8list[[i]], conf.level = 0.95)$conf.int
  moy <- mean(bdd8list[[i]])
  qq <- quantile(bdd8list[[i]], probs = c(0.25, 0.75))
  return(c(moy, ci, qq))
})
colnames(resCI95) <- names(bdd8list)
rownames(resCI95) <- c("moy", "CI-", "CI+", "q1", "q3")

Etape 05 : quartiles

print(resCI95)

##              1          2         3         4         5        6        7
## moy  0.5091296  0.7647209  7.421766  8.188332 11.196756 17.16307 18.49170
## CI-  0.3538382  0.6214972  7.198553  7.958890 11.028022 16.93950 18.34085
## CI+  0.6644210  0.9079446  7.644980  8.417774 11.365490 17.38664 18.64254
## q1  -2.2337695 -1.4388965  4.099141  4.857246  8.310664 13.39904 16.09113
## q3   3.4823438  1.9169629 10.292109 10.785469 14.162422 20.33674 20.34094
##            8         9       10       11        12
## moy 15.33637 11.900627 5.383014 3.891985  2.825919
## CI- 15.20963 11.763177 5.044672 3.709193  2.607323
## CI+ 15.46312 12.038078 5.721356 4.074776  3.044515
## q1  13.10656  9.904316 1.885664 0.289668 -1.828740
## q3  17.25197 13.826484 8.271602 7.057832  7.134541

Etape 06 : rep. graph.

par(mar = c(4, 4, 1, 1))
plot(x = 1:12, y = resCI95[1, ], ylim = c(0, 25),
  xlab = "Month", ylab = "Temperature from 8:00 to 8:59 (°C)",
  type = "l", lwd = 2, axes = FALSE, panel.first = {
    grid()
    polygon(
      x = c(1:12, 12:1), 
      y = c(resCI95[4, ], rev(resCI95[5, ])), 
      col = "salmon", border = NA)
    polygon(
      x = c(1:12, 12:1), 
      y = c(resCI95[2, ], rev(resCI95[3, ])), 
      col = 2, border = NA)
  })
axis(1, at = 1:12, labels = month.abb, las = 3)
axis(2)
text(x = 1:12, y = resCI95[1, ], 
  labels = round(resCI95[1, ]), 
  pos = c(1, 1, 2, 4, 4, 2, 3, 1, 4, 1, 1, 1))

Etape 06 : rep. graph.

Etape 06 : rep. graph.

cBar <- barplot(resCI95[1, ], axes = FALSE, 
  names.arg = "", ylim = c(0, 28))
axis(1, at = cBar, labels = month.abb, las = 3)
axis(2)
arrows(
  cBar, resCI95[2, ], 
  cBar, resCI95[3, ], 
  lwd = 1.5, angle = 90,
  code = 3, length = 0.05)
arrows(
  cBar, resCI95[4, ], 
  cBar, resCI95[5, ], 
  lwd = 1.5, angle = 90,
  code = 3, length = 0.05)

Etape 06 : rep. graph.

Etape 06 : rep. graph.

myColP100 <- round((resCI95[1, ] - min(resCI95[1, ])) / 
  (max(resCI95[1, ]) - min(resCI95[1, ])) * 100 + 1)
myPal <- colorRampPalette(c("blue", "red"))(101)
myCol <- myPal[myColP100]
cBar <- barplot(resCI95[1, ], axes = FALSE, 
  names.arg = "", ylim = c(0, 28), col = myCol)
axis(1, at = cBar, labels = month.abb, las = 3)
axis(2)
arrows(
  cBar, resCI95[2, ], 
  cBar, resCI95[3, ], 
  lwd = 1.5, angle = 90,
  code = 3, length = 0.05)
arrows(
  cBar, resCI95[4, ], 
  cBar, resCI95[5, ], 
  lwd = 1.5, angle = 90,
  code = 3, length = 0.05)

Etape 06 : rep. graph.

Etape 06 : rep. graph.

myColP100 <- round((resCI95[1, ] - min(resCI95[1, ])) / 
  (max(resCI95[1, ]) - min(resCI95[1, ])) * 100 + 1)
myPal <- colorRampPalette(c("blue", "red"))(101)
myCol <- myPal[myColP100]
cBar <- barplot(resCI95[1, ], axes = FALSE, 
  names.arg = "", ylim = c(0, 28), col = myCol)
abline(h = axTicks(2), col = "lightgray", lty = 2)
axis(1, at = cBar, labels = month.abb, las = 3)
axis(2)
arrows(
  cBar, resCI95[2, ], 
  cBar, resCI95[3, ], 
  lwd = 1.5, angle = 90,
  code = 3, length = 0.05)
arrows(
  cBar, resCI95[4, ], 
  cBar, resCI95[5, ], 
  lwd = 1.5, angle = 90,
  code = 3, length = 0.05)

Etape 06 : rep. graph.

Etape 06 : rep. graph.

# error bars
resCI95 <- sapply(seq_along(bdd8list), function(i){
  ci <- t.test(bdd8list[[i]], conf.level = 0.95)$conf.int
  moy <- mean(bdd8list[[i]])
  qq <- quantile(bdd8list[[i]], probs = c(0.25, 0.75))
  se <- sd(bdd8list[[i]]) / sqrt(length(bdd8list[[i]]))
  return(c(moy, ci, qq, se))
})
colnames(resCI95) <- names(bdd8list)
rownames(resCI95) <- c("moy", "CI-", "CI+", "q1", "q3", "se")
myColP100 <- round((resCI95[1, ] - min(resCI95[1, ])) / 
  (max(resCI95[1, ]) - min(resCI95[1, ])) * 100 + 1)
myPal <- colorRampPalette(c("blue", "red"))(101)
myCol <- myPal[myColP100]
cBar <- barplot(resCI95[1, ], axes = FALSE, 
  names.arg = "", ylim = c(0, 28), col = myCol)
abline(h = axTicks(2), col = "lightgray", lty = 2)
axis(1, at = cBar, labels = month.abb, las = 3)
axis(2)
arrows(
  cBar, resCI95[1, ] + resCI95[6, ]*2, 
  cBar, resCI95[1, ] - resCI95[6, ]*2, 
  lwd = 1.5, angle = 90,
  code = 3, length = 0.05)

Etape 06 : rep. graph.