Exercise 4.1
- Insert in a Markdown document the correlation matrix of the numeric variables of the iris data.
library("kableExtra")
vs_dt <- as.data.frame(round(cor(iris[, 1:4]), 2))
vs_dt[1:4] <- lapply(vs_dt[1:4], function(x) {
cell_spec(x, bold = T, color = spec_color(abs(x), scale_from = c(-1, 1)),
font_size = spec_font_size(abs(x), scale_from = c(0, 1)))
})
kbl(vs_dt, escape = F, align = "c") %>% kable_classic("striped", full_width = F)
|
|
Sepal.Length
|
Sepal.Width
|
Petal.Length
|
Petal.Width
|
|
Sepal.Length
|
1
|
-0.12
|
0.87
|
0.82
|
|
Sepal.Width
|
-0.12
|
1
|
-0.43
|
-0.37
|
|
Petal.Length
|
0.87
|
-0.43
|
1
|
0.96
|
|
Petal.Width
|
0.82
|
-0.37
|
0.96
|
1
|
- Insert the table of results of a regression analysis of the iris data
mod_full <- lm(Sepal.Length ~ ., data = iris)
stargazer::stargazer(mod_full, type = "html",
title = "Regression results", header = F)
Regression results
|
|
|
|
Dependent variable:
|
|
|
|
|
|
Sepal.Length
|
|
|
|
Sepal.Width
|
0.496***
|
|
|
(0.086)
|
|
|
|
|
Petal.Length
|
0.829***
|
|
|
(0.069)
|
|
|
|
|
Petal.Width
|
-0.315**
|
|
|
(0.151)
|
|
|
|
|
Speciesversicolor
|
-0.724***
|
|
|
(0.240)
|
|
|
|
|
Speciesvirginica
|
-1.023***
|
|
|
(0.334)
|
|
|
|
|
Constant
|
2.171***
|
|
|
(0.280)
|
|
|
|
|
|
|
Observations
|
150
|
|
R2
|
0.867
|
|
Adjusted R2
|
0.863
|
|
Residual Std. Error
|
0.307 (df = 144)
|
|
F Statistic
|
188.251*** (df = 5; 144)
|
|
|
|
Note:
|
p<0.1; p<0.05; p<0.01
|
Exercise 4.2
- Find the code which permits to obtain with ggplot2 this figure:
op <- par(oma = c(1, 1, 0, 1), las = 1)
boxplot(Sepal.Length ~ Species, data = iris)
points(as.numeric(iris$Species) + rnorm(150, 0, 0.1), iris$Sepal.Length)
points(c(1, 2, 3), tapply(iris$Sepal.Length, iris$Species, mean),
col = "red", pch = 16, cex = 2)
par(op)
library(ggplot2)
ggplot(iris, aes(x = Species, y = Sepal.Length)) +
geom_boxplot()+
geom_jitter(position=position_jitter(0.2)) +
stat_summary(fun=mean, geom="point", shape=20, size=14, color="red", fill="red")
- Find the code in R base code which permits to obtain this figure:
data("diamonds")
ggplot(diamonds,
aes(x = carat,
y = price)) +
geom_point() +
ggtitle("My scatter plot")
par(las = 1, cex.axis = 0.8, cex.lab = 0.8)
plot(price ~ carat, data = diamonds, pch = 16, cex = 0.7, xlab = "carat",
ylab= "prix", main = "Scatter plot")
abline(h = seq(0, 20000, by = 5000), v = seq(0, 4, by = 0.5), col = "lightgray", lty = "dotted")
Exercise 4.3
- On the lung data used previously, make a mosaic plot between status and sex variable.
library("survival")
data(lung)
## Warning in data(lung): data set 'lung' not found
tab <- xtabs(~ status + sex, lung)
vcd::mosaic(tab, shade = TRUE, legend = TRUE)
- On the lung data, make a ridge plot of variable age with respect to status.
library(ggridges)
ggplot(lung) +
aes(x = age, y = factor(status), fill = factor(status)) +
geom_density_ridges() +
theme_ridges() +
labs("Age by death/live") +
theme(legend.position = "none")
## Picking joint bandwidth of 3.17
- Make a correlation plot of variables ph.karno, pat.karno, meal.cal, wt.loss in the lung data.
library(ggcorrplot)
r <- cor(lung[, 7:10], use = "complete.obs")
ggcorrplot(r, hc.order = TRUE, type = "lower", lab = TRUE)
