Chapter 9 exercise
Aardvark
5th Dec
Sheff_temp <- read.csv("Sheffield_max_temp_data.csv", header = FALSE)
colnames (Sheff_temp) <- c ("Year", "Month", "Max_temp")Exercise 9.1.
boxplot (Sheff_temp$Max_temp ~ Sheff_temp$Month,
xlab = "", ylab = "Celsius", xaxt = "n",
main = "Average maximum monthly temperature in Sheffield, 1883 - 2014")
axis(side = 1, at = 1:12, labels = month.abb,
las = 3, font =3)
Exercise 9.2.
Sheff_temp_noNA <- na.omit(Sheff_temp)
boxplot (Sheff_temp_noNA$ Max_temp ~ Sheff_temp_noNA$ Month,
xlab = "", ylab = "Celsius", xaxt = "n",
main = "Average maximum monthly temperature in Sheffield, 1883 - 2014")
axis(side = 1, at = 1:12, labels = month.abb,
las = 3, font =3)
max_temperature <- which.max (Sheff_temp_noNA $Max_temp)
max_temperature_row <- Sheff_temp_noNA [max_temperature,, ]
max_temperature_row## Year Month Max_temp
## 1483 2006 7 25.6min_temperature <- which.min (Sheff_temp_noNA $Max_temp)
min_temperature_row <- Sheff_temp_noNA [min_temperature,, ]
min_temperature_row## Year Month Max_temp
## 770 1947 2 -0.6points (max_temperature_row$Month, max_temperature_row$Max_temp ,
pch = 20, col = "red")
points (min_temperature_row$Month, min_temperature_row$Max_temp ,
pch = 20, col = "dodgerblue")
text(3.5, min_temperature_row$Max_temp,
"The lowest temperature",
font = 3, col = "dodgerblue", cex = 0.6)
text(5.5 , max_temperature_row$Max_temp,
"The highest temperature",
font = 3, col = "red", cex = 0.6)
Exercise 9.3.
par (mfrow = c (1,1))
boxplot (Sheff_temp$Max_temp ~ Sheff_temp$Month,
xlab = "Months", ylab = "Celsius", xaxt = "n",
main = "Average maximum monthly temperature \n in Sheffield, 1883 - 2014",
col = c (rep ( c("dodgerblue", "indianred1", "green3","goldenrod1") , c (2, 3, 3, 3) )),
cex.main = 0.9
)
axis(side = 1, at = 1:12, labels = month.abb,
las = 3, font =3)
abline (v = c (1:12), col = "grey", lty = 2)
abline (h = c (seq(5 , 25, 5)), col = "grey")
Sheff_data <- data.frame (Sheff_temp_noNA)
plot (Sheff_temp_noNA[Sheff_temp_noNA$Month == 1,]$Year,
Sheff_temp_noNA[Sheff_temp_noNA$Month == 1,]$Max_temp,
type = "l",
col = "blue",
xlab = "year",
ylab = "Average Temperature (°C)",
main = "Average maximum monthly temperature \n in Sheffield, 1883 - 2014",
cex.main = 0.9)
par (mfrow = c (1,1))- Missing data: omit using na.omit() function
- Not helpful to have the same scale, and there appears to be no trend in the second plot.
Exercise 9.4.
qanda <- read.csv ("selective_affinities_aftermodified.csv")
attach (qanda)
data_frame_9.4 <- data.frame(Handspan, Siblings, Fish, Height)
pairs (data_frame_9.4)
detach (qanda)
par(mfrow = c (1,1))- People with greater heights tend to have larger handspan.
- People eat fish tend to have higher height, and greater handspan.
Exercise 9.5.
mod.squ <- function (a,d) { # the function takes two numbers, a and d,
f <- abs (a^2 - d^2) # creates the value, f , we cwant to calculate
f # then outputs f
}
mod.squ(a = 3, d = 6)## [1] 27aaaa <- 2
dddd <- 4
mod.squ (a = aaaa, d = dddd)## [1] 12vector_1 <- c (1, 5, 6, 10)
vector_1## [1] 1 5 6 10vector_2 <- c (2, 0, 6, 3)
vector_2## [1] 2 0 6 3mod.squ(a = vector_1, d = vector_2)## [1] 3 25 0 91matrix_1 <- matrix (data = c (1, 4, 6, 2, 8, 15, 2, 5, 6), nrow = 3)
matrix_1## [,1] [,2] [,3]
## [1,] 1 2 2
## [2,] 4 8 5
## [3,] 6 15 6matrix_2 <- matrix (data = c (1, 8, 5, 9, 24, 7, 3, 4, 12), nrow = 3)
matrix_2## [,1] [,2] [,3]
## [1,] 1 9 3
## [2,] 8 24 4
## [3,] 5 7 12mod.squ(a = matrix_1, d = matrix_2)## [,1] [,2] [,3]
## [1,] 0 77 5
## [2,] 48 512 9
## [3,] 11 176 108The position of each output, if putting an input with a vector or a matrix, will be the corresponded to the position of two input vector / matrix with same size.
Exercise 9.6.
Sheff.mth.plot <- function (mth) { # use a number for the month, "mth"
mth_data <- Sheff_temp_noNA[Sheff_temp_noNA[,2] == mth, ]
plot(x = mth_data$Year, y = mth_data$Max_temp, type = "l",
main = month.name[mth], xlab = "year", ylab = "temperature")
}
par(mfrow = c (3, 1))
Sheff.mth.plot( 2 )
Sheff.mth.plot( 5 )
Sheff.mth.plot( 8 )
Exercise 9.7.
library(plotrix)
par(mfrow = c (1,1))
Trantor_Uni <- c (87, 88, 87, 88, 88, 89, 87)
plot (Trantor_Uni, ##### This is the name of the data we want to plot
ylab = "percentage", ##### This is the name/label of the y-axis
ylim = c (80, 92),
yaxt = "n",
xlab = "Satisfaction %", ##### Similarly, the name/label of the x-axis
col = "red", ##### The color of the plot
pch = ".", ##### This is the type of symbol of dots, see R reference Card P3 top right corner
type = "o", ##### This specifies the type of plot, see R reference Card P3 plotting section down left corner
lty = 1, lwd = 3, ##### These two controls the type of lines and width of lines, see P3 top right corner
main = "Trantor University student satisfaction")
axis(2, at = 80:92)
axis.break (2, breakpos = 80, style = "zigzag")
Exercise 9.8.
the purpose of a violin plot: It is used to visualize the distribution of numerical data.
attach(Sheff_temp_noNA)
library(ggplot2)
ggplot (Sheff_temp_noNA,
aes (x = factor (Month, labels = month.abb),
y = Max_temp, fill = factor (Month)
) ##### Maps the month to the x-axis,
##### mean temperature to the y-axis,
##### and uses the fill aesthetic for different months.
)+
geom_violin (trim = FALSE, alpha = 0.8) + ### Creates a violin plot.
### trim: controls whether to trim the width of the violin
### alpha value to control the transparency of the violins according to your preferences
### higher value of alpha -> less transparent
scale_x_discrete(name = "Month") + ### Sets the axis labels.
scale_y_continuous(name = "Mean Temperature") +
ggtitle("Violin plot of Monthly Mean Temperature of Sheffield")+ ### Gives the plot title.
theme_minimal()+ ### Sets the plot theme to a minimal style.
theme(panel.background = element_rect(fill = "pink")) ### Sets the background color of the plot to pink
detach(Sheff_temp_noNA)Violin plot can:
- Show asymmetry and skewness in the data through the shape of the violin
- Provides a continuous representation of the data distribution.
Exercise 9.9.
See Exercise 9.3 on the same document, it has been improved.
Exercise 9.10.
The plot for exercise 6.4. Cocoa Prices from 2012 to 2014
# add the price and mean lines for all three years using
# the a and b arguments for abline(), legend, text and axis features.
cocoa <- read.csv ("Cocoa_prices.csv", header = FALSE)
Mths <- substr(cocoa$V1, start = 1, stop = 3)
Yrs <- paste ("20", substr ( cocoa$V1, start = 5, stop = 6))
cocoa_updated <- data.frame (Yrs, Mths, cocoa$V2)
Price <- cocoa_updated$cocoa.V2
cocoa_updated_price <- data.frame (Yrs, Mths, Price)
cocoa_wide <- reshape ( cocoa_updated_price, timevar = "Yrs", idvar = "Mths", direction = "wide")
colnames (cocoa_wide) <- c ( colnames (cocoa_wide)[1], "2012", "2013", "2014")
##### plot the first line of year 2012
plot(cocoa_wide$"2012",
xlab = "months", ylab = "price ($)", ylim = c (2100, 3300),
main = "Cocoa Prices from 2012 to 2014",
col = "black",
xaxt='n', bty = "l")
months <- as.character( cocoa_wide$Mths)
axis (side = 1, at = 1:12, labels = months)
lines(cocoa_wide$"2012", col = "black", lwd= 3)
abline ( h = mean (cocoa_wide$"2012"),
lty = 3, lwd = 2,
col = "black")
# plot the second line of year 2013
points(cocoa_wide$"2013", col = "deepskyblue", pch = 15)
lines(cocoa_wide$"2013", col = "deepskyblue", lty = 1, lwd = 3)
abline ( h = mean (cocoa_wide$"2013"),
lty = 3, lwd = 2,
col = "blue")
# plot the last line of the year 2014
points(cocoa_wide$"2014", col = "red", pch = 20)
lines(cocoa_wide$"2014", col = "red", lty = 1, lwd = 3)
abline ( h = mean (cocoa_wide$"2014"),
lty = 3, lwd = 2,
col = "red")
# add the legend and text
legend ( x = 10, y = 2300, legend = c ("2012", "2013", "2014"),
col = c ( "black", "blue", "red"),
lty = rep ( 1 , times = 3), cex = 0.8)
text (x = 3, y = 2700,
labels = paste ("The dotted lines are annual means",
sep = ""),
font = 3, col = "black")
# add means
text (x = 10, y = mean(cocoa_wide$"2012") -50,
labels = substitute (paste ("mean:",
# Substitute() function could be used to show greek letters,
# but leaves alone any text in quotation marks.
mu[ 2012], # The [] here is completely different from the one we have used so far,
# They tell the substitute() function to display whatever is inside them
# as a subscript (^ for superscripts)
" = $",
a),
list (a = round (mean (cocoa_wide$"2012"),
0),
sep = "")
),
font = 3, col = "black", cex = 0.8)
text (x = 4, y = mean(cocoa_wide$"2013") +50,
labels = substitute (paste ("mean:", mu[ 2013], " = $", a),
list (a = round (mean (cocoa_wide$"2013"),
0),
sep = "")
),
font = 3, col = "blue", cex = 0.8)
text (x = 8, y = mean(cocoa_wide$"2014") +50,
labels = substitute (paste ("mean:", mu[ 2014], " = $", a),
list (a = round (mean (cocoa_wide$"2014"),
0),
sep = "")
),
font = 3, col = "red", cex = 0.8)
- don’t forget to count how many opening and closing brackets there are to avoid silly errors.
Exercise 9.11.
hist (qanda$Handspan, prob = TRUE, col = "lightblue",
main = "Handspan - default bins \n bandwidth = 2", xlab = "handspan", ylab = "frequency")
lines (density(qanda$Handspan), col = "red", lwd = 2)
hist (qanda$Handspan, prob = TRUE, breaks = 8/0.5, col = "lightblue",
main = "Handspan - bandwidth = 1", xlab = "handspan", ylab = "frequency")
lines (density(qanda$Handspan, adjust = 0.5), col = "red", lwd = 2)
hist (qanda$Handspan, prob = TRUE, breaks = 8/0.25, col = "lightblue",
main = "Handspan - bandwidth = 0.5", xlab = "handspan", ylab = "frequency")
lines (density(qanda$Handspan, adjust = 0.25), col = "red", lwd = 2)
Exercise 9.12
We can use a Mosaic plot:
# Load the dataset
data("HairEyeColor")
HairEyeColor_df <- as.table(HairEyeColor)
par(mfrow = c (1,1))
### Create a mosaic plot
mosaicplot(Freq ~ Hair + Eye + Sex,
data = HairEyeColor_df,
color = c ("grey100", "grey80", "grey60", "grey40", "grey20", "black"),
main = "HairEyeColor Dataset")