Data Visualization - Lesson 5: Visualizing data in R

Downloading packages

packages <- c('dplyr', 'ggplot2', 'readxl', 'readr', 'tidyverse', 'ggthemes', 'knitr', 'extrafont', 'dplyr', 'scales', 'lubridate', 'gghighlight') 
# Checking for package installations on the system and installing if not found.
if (length(setdiff(packages, rownames(installed.packages()))) > 0) {
  install.packages(setdiff(packages, rownames(installed.packages())))  
}
# Packages to use
for(package in packages){
  library(package, character.only = TRUE)
}

Downloading and cleaning data

u <- "bikesharedailydata.csv"
data <- as.data.frame(read_csv(u, col_names = TRUE))
data <- data%>%mutate(date = as.Date(dteday, "%m/%d/%Y"))
data <- data%>%mutate(Month = format(date,"%B"), year = format(date, "%Y"), Week_day = weekdays(date))

data2<- data%>%dplyr::select(date, casual, registered)
data2<- gather(data2, rental_type, count, casual:registered, factor_key=TRUE)

data3<- data%>%dplyr::select(date, casual, registered, cnt)%>%mutate(p_casual = casual/cnt, p_registered = registered/cnt)%>%dplyr::select(date, p_casual, p_registered)
colnames(data3) <- c("date", "casual", "registered")
data3<- gather(data3, rental_type, count, casual:registered, factor_key=TRUE)

Bar Chart

data1 <- data%>%filter(year == "2012")%>%mutate(High = ifelse(Month == "August"|Month == "September", "Yes", "No"))
ggplot(data = data1, aes(x= factor(Month, levels = month.name) ,y=cnt, fill = High)) +
  geom_bar(stat="identity") +  theme_bw() + 
  scale_fill_manual(values = c( "Yes"="#00FA9A", "No"="#C0C0C0" ), guide = FALSE ) +
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(title = "Bicycle rental by month in 2012", x = "Months", y = "Count",
       subtitle = "August and September are the months with the largest number of bicycle rentals in 2012",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)")  + scale_y_continuous(labels = comma) 

Line Charts

data1 <- data%>%filter(year == "2012") 
ggplot(data = data1, aes(x= date ,y=casual)) +
  geom_line(color="#00BED8", size = 1) +  geom_point(color="#00BED8", size = 2)  + theme_bw() + 
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(title = "Daily rental of casual users in 2012", x = "Date", y = "Rentals", subtitle = "There is a periodic pattern in the days and a seasonal variation which increases with the level of the series",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)")  + scale_y_continuous(labels = comma)

data1 <- data2%>%filter(year == "2012") 
ggplot(data = data1, aes(x= date ,y=count, color = rental_type)) +
  geom_line( size = 1) +  geom_point( size = 2)  + theme_bw() + 
   scale_color_manual(values = c("#00BED8","#696969")) +
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(color = "Rental type",title = "Daily bicycle rental of casual and registered users in 2012", x = "Date", y = "Rentals", subtitle = "There is a periodic pattern in the days for casual and registered bicycle users",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)")  + scale_y_continuous(labels = comma) +
  facet_wrap(~rental_type, ncol = 1)

Area Chart

data1 <- data%>%filter(year == "2012") 
ggplot(data = data1, aes(x= date ,y=temp)) +
  geom_area(color="#3CB371", fill = "#3CB371")  + theme_bw() + 
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(title = "Daily temperature in 2012", x = "Date", y = "Temperature", subtitle = "July registered the highest temperature over 75F",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)")  

Staked Area Charts

data2 <- data2%>%mutate(year = format(date, "%Y"))%>%filter(year == "2012") 
data2$Rental_type <- factor(data2$rental_type, levels = c('registered', 'casual'))
ggplot(data = data2, aes(x= date ,y=count)) +
  geom_area(aes(fill = Rental_type), position = 'stack')  + theme_bw() + 
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(title = "Bicycle rental by rental type and date in 2012", x = "Date", y = "Rentals", fill = "Rental Type", subtitle = "There were more bicycle registered users than casual users every day in 2012",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)" )  + scale_y_continuous(labels = comma) + scale_fill_manual(values=c("#A9A9A9", "#00BED8")) 

data3 <- data3%>%mutate(year = format(date, "%Y"))%>%filter(year == "2012") 
data3$Rental_type <- factor(data3$rental_type, levels = c('casual', 'registered'))
ggplot(data = data3, aes(x= date ,y=count)) +
  geom_area(aes(fill = Rental_type), position = 'stack')  + theme_bw() + 
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(title = "Bicycle rental by rental type and date in 2012", x = "Date", y = "Rentals", fill = "Rental Type", subtitle = "There were more bicycle registered users than casual users every day in 2012",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)")  + scale_y_continuous(labels = scales::percent) + scale_fill_manual(values=c("#A9A9A9", "#00BED8")) 

Histogram

data1 <- data%>%filter(year == "2012")
ggplot(data = data1, aes(x=cnt)) +
  geom_histogram(binwidth = 300, fill="#00BED8", color = "white") +  theme_bw() + 
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(title = "Frequency distribution of registered bicycle rentals in 2012", x = "Rentals", y = "Frequency", subtitle = "The graph indicates that the distribution of the data is negatively skewed",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)")  + scale_y_continuous(labels = comma) + scale_x_continuous(labels = comma)

Density Plot

data1 <- data%>%filter(year == "2012")
ggplot(data = data1, aes(x=registered)) +
  geom_density(fill="#00BED8", color = "#00BED8") +  theme_bw() + 
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(title = "Probability densities and frequency distribution of registered bicycle rentals in 2012",x = "Rentals", y = "Density", subtitle = "The graph indicates that the distribution of the data is negatively skewed",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)") + scale_x_continuous(labels = comma)  + scale_y_continuous(labels = scales::number_format(accuracy = 0.00001, decimal.mark = ',')) + geom_vline(aes(xintercept=round(mean(data$cnt),2)),
            color="#696969", linetype="dashed", size=1)

Boxplot

data1 <- data%>%filter(year == "2012") 
ggplot(data = data1, aes(x= factor(Month, levels = month.name) ,y=cnt)) +
  geom_boxplot(color="#00BED8", size = 1) +  theme_bw() + 
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(title = "Boxplot of bicycle rentals by month in 2012", x = "Months", y = "Rentals", subtitle = "Rentals at the end of the year have more variability. October and April have days with very few rentals",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)")  + scale_y_continuous(labels = comma)

NA
NA

ScatterPlots

data1 <- data%>%mutate(Temperature = ifelse(temp>0.8, "Temperature > 0.8", "Temperature <= 0.8")) 
ggplot(data = data1, aes(y= temp ,x=cnt, color = Temperature)) +
  geom_point(size = 4) +  theme_bw() + 
  scale_color_manual(values = c( "#C0C0C0", "#3CB371")) +  geom_smooth(method=lm) +
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(color = " ", title = "Relationship between temperature and bicycle rentals",x = "Rentals", y = "Temperature", 
subtitle = "There is a positive relationship between temperature and rides, however, when the temperature peaks above 80 degrees, there is a decrease",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)") + scale_x_continuous(labels = comma)

NA
NA

ggplot(data = data, aes(y= windspeed ,x=cnt)) +
  geom_point(color="#588BAE", size = 4) +  theme_bw() + geom_smooth(method=lm) +
  theme(
        text=element_text(size=17,  family="Comic Sans MS"),
        panel.border = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor =  element_blank(),
        axis.line = element_line(color = "gray"),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()) + labs(title = "Relationship between humidity and bicycle rentals",x = "Rentals", y = "Humidity", 
subtitle = "There is a negative relationship between humidity and bicycle rides",
                                               caption =  "Jose Vilardy | Source = Fanaee-T, Hadi, and Gama, J. (2013)") + scale_x_continuous(labels = comma)

NA
NA