R project

Arthur Martins Ferreira de Sousa

Visualizing data and Boxplots

Importing libraries

library(dplyr)
library(tidyverse)
library(reshape2)
library(plotly)
library(kableExtra)
library(knitr)

Reading data-set

setwd("C:/Users/arthu/OneDrive/Área de Trabalho/R_projects/trabalhos de VED/Databases")
getwd()
data = readxl::read_xlsx("preço_hotel.xlsx")
df = data.frame(cidade = data$Cidade, duas_estrelas = data$`Duas-Estrelas`,
                tres_estrelas = data$`Três-Estrelas`, quatro_estrelas = data$`Quatro-Estrelas`)

Selecting and arranging specif columns in order from data-set

Two-star hotels

o1 = select(df, cidade, duas_estrelas)
o2 = arrange(o1, duas_estrelas)

Two-star Hotels

City	Price (£)
Shanghai	22
Bangkok	23
Nova Delhi	29
Benidorm	32
Budapeste	33
Las Vegas	33
Cancun	38
Orlando	38
Taline	40
Hong Kong	43
Mumbai	43
Viena	43
Dublin	44
Pequim	45
Istanbul	47
Lisboa	48
Dubai	50
Tóquio	52
Berlim	54
Sydney	55
Madri	56
Singapura	56
Oslo	60
Helsinki	62
Los Angeles	63
Barcelona	65
Nice	66
Londres	67
Munique	67
Edinburgo	68
Miami	68
Estocolmo	69
Jerusalém	70
Frankfurt	71
Montreal	71
Toronto	71
Bruxelas	73
Chicago	73
Copenhagen	73
São Francisco	73
Vancouver	73
Roma	75
Seattle	75
Amsterdam	77
Cidade do Cabo	77
Paris	78
Washington	80
Veneza	82
Zurique	84
Boston	90
Geneva	106
Nova York	110

Three-star hotels

o3 = select(df, cidade, tres_estrelas)
o4 = arrange(o3, tres_estrelas)

Four-star hotels

o5 = select(df, cidade, quatro_estrelas)
o6 = arrange(o5, quatro_estrelas)

R base Boxplot

boxplot(df$duas_estrelas, df$tres_estrelas, df$quatro_estrelas,
        main = "Boxplots", names = c("2-stars", "3-stars", "4-stars"), 
        col = c("yellow","orange","red"), ylab = "Price (£)", xlab = "Hotels",
        whisklty = 5, whisklwd = 1.5, staplelwd = 1.75)

ggplot2 Boxplot

#Drawing multiple plots with the ggplot2 requires data in long format --> (reshape2 package)
df_long = melt(df)

p = ggplot(df_long, aes(x = variable, y = value))+
  geom_boxplot(color = "black", fill = "red", alpha = 0.4, outlier.color = "red", outlier.alpha = 1)+
  stat_summary(fun.y = mean, geom = "point", shape = 20, size = 3, color = "red")+
  labs(title = "Boxplots",x = "Hotels", y = "Price (£)")+
  scale_x_discrete(labels = c("two stars", "three stars", "four stars"))+
  theme(plot.title = element_text(hjust = 0.5), plot.tag.position = "topright")
ggplotly(p)