Actividad 3
Ashwin Malshe
02 October 2020
Get the data
top10 <- readRDS(gzcon(url("https://github.com/ashgreat/datasets/blob/master/top10_billionaires.RDS?raw=true"))) %>%
filter(year != 2018)Take a peek
head(top10, 10)## # A tibble: 10 x 6
## name lastName netWorth rank position year
## <chr> <chr> <dbl> <int> <int> <dbl>
## 1 Jeff Bezos Bezos 112000 1 1 2017
## 2 Bill Gates Gates 90000 2 2 2017
## 3 Warren Buffett Buffett 84000 3 3 2017
## 4 Bernard Arnault Arnault 72000 4 4 2017
## 5 Mark Zuckerberg Zuckerberg 71000 5 5 2017
## 6 Amancio Ortega Ortega 70000 6 6 2017
## 7 Carlos Slim Helu Slim Helu 67100 7 7 2017
## 8 Charles Koch Koch 60000 8 8 2017
## 9 Larry Ellison Ellison 58500 10 10 2017
## 10 Michael Bloomberg Bloomberg 50000 11 11 2017The following code will give error. Why?
ggplot(top10, aes(x = name, y = netWorth, fill = year)) +
geom_bar() +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth")Adding identity (or you can use geom_col)
ggplot(top10, aes(x = name, y = netWorth, fill = year)) +
geom_bar(stat = "identity") +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth")
The position of the bars is stacked. We want them to be dodged
Play around with the widths
ggplot(top10, aes(x = name, y = netWorth, fill = year)) +
geom_bar(stat = "identity",
position = position_dodge(0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth")
Why is the dodging not working?
In order to godge the bars, the fill argument should be mapped to a categorical variable. So I will create a new categorical variable.
top10$year2 <- factor(top10$year)ggplot(top10, aes(x = name, y = netWorth, fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth")
Change the legend title to “Year”
ggplot(top10, aes(x = name,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth")
The names appear in alphabetical order. Either we can recode the names to be factors or we can change the ordering based on the ranking in ggplot.
ggplot(top10, aes(x = reorder(name, rank),
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = "Year")
Something is still missing here. Any idea what’s going on?
top2017 <- top10 %>%
filter(year == 2017) %>%
select(name, rank) %>%
rename(rank2017 = rank)
top10 <- inner_join(top10, top2017, by = "name")
# print column names
colnames(top10)## [1] "name" "lastName" "netWorth" "rank" "position" "year" "year2"
## [8] "rank2017"top10$name <- reorder(top10$name, top10$rank2017)Also, split the names on two lines so that we can render them nicely
top10$name2 = gsub(" ", " \n", top10$name)
top10$name2 = reorder(top10$name2, top10$rank2017)ggplot(top10, aes(x = name2,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = "Year")
Adjust the X axis ticks by changing the arguments in element_text()
ggplot(top10, aes(x = name2,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
theme(axis.text.x = element_text(angle = , hjust =, size = ))
Add commas to Y axis labels
ggplot(top10, aes(x = name2,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
theme(axis.text.x = element_text(angle = 45,
hjust = 0.6)) +
scale_y_continuous(labels = scales::comma)
Label the bars
ggplot(top10, aes(x = name,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
scale_y_continuous(labels = scales::comma) +
geom_text(aes(label = netWorth)) +
theme(axis.text.x = element_text(angle = 45,
hjust = 0.6))
The labels are clearly badly rendered We need to dodge them the same way we dodged the bars
ggplot(top10, aes(x = name,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
theme(axis.text.x = element_text(angle = 45,
hjust = 0.6)) +
scale_y_continuous(labels = scales::comma) +
geom_text(aes(label = netWorth),
position = position_dodge(width = 0.9))
Looks better but still not the best Rotate the labels by 90 degrees
ggplot(top10, aes(x = name, y = netWorth, fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
theme(axis.text.x = element_text(angle = 45,
hjust = 0.6)) +
scale_y_continuous(labels = scales::comma) +
geom_text(aes(label = netWorth),
position = position_dodge(width = 0.9),
angle = 90)
Push the labels inside the bars
ggplot(top10, aes(x = name,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
theme(axis.text.x = element_text(angle = 45,
hjust = 0.6)) +
scale_y_continuous(labels = scales::comma) +
geom_text(aes(label = netWorth),
position = position_dodge(width = 0.9),
angle = 90,
hjust = 1.1)
Change the color of the labels
ggplot(top10, aes(x = name,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
theme(axis.text.x = element_text(angle = 45,
hjust = 0.6)) +
scale_y_continuous(labels = scales::comma) +
geom_text(aes(label = netWorth),
position = position_dodge(width = 0.9),
angle = 90,
hjust = 1.1,
color = "white")
Make this look cool using premade themes
Go to the top of the file and insert library(ggthemes) in the setup chunk.
ggplot(top10, aes(x = name2,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = NULL,
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
theme(axis.text.x = element_text(angle = 45,
hjust = 0.6)) +
scale_y_continuous(labels = scales::comma) +
geom_text(aes(label = netWorth),
position = position_dodge(width = 0.9),
angle = 90,
hjust = 1.1,
color = "white") +
theme_economist()
Usually themes have their own parameters and they can override our themes. So let’s put our theme the last.
ggplot(top10, aes(x = name2,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = NULL,
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
scale_y_continuous(labels = scales::comma) +
geom_text(aes(label = netWorth),
position = position_dodge(width = 0.9),
angle = 90,
hjust = 1.1,
color = "white") +
theme_economist() +
theme(axis.text.x = element_text(angle = 45,
hjust = 0.5,
vjust = 0.7))
Save the file to your hard disk
ggsave("Top_10_Billionaires.png",
width = 9,
height = 6)Try a different theme
ggplot(top10, aes(x = name2,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = NULL,
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
scale_y_continuous(labels = scales::comma) +
geom_text(aes(label = netWorth),
position = position_dodge(width = 0.9),
angle = 90,
hjust = 1.1,
color = "white") +
theme_fivethirtyeight() +
theme(axis.text.x = element_text(angle = 0,
hjust = 0.5,
vjust = 0.7))
Save the file to your hard disk
ggsave("Top_10_Billionaires_2.png",
width = 9,
height = 6,
dpi=900)Are you nostalgic about Excel graphs?
ggplot(top10, aes(x = name,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
labs(x = "Name",
y = "Net Worth in Million $",
title = "Top 10 Billionaires' Net Worth",
fill = NULL) +
scale_y_continuous(labels = scales::comma) +
geom_text(aes(label = netWorth),
position = position_dodge(width = 0.9),
angle = 90,
hjust = 1.1,
color = "white") +
theme_excel() +
theme(axis.text.x = element_text(angle = 45, hjust = 0.5, vjust = 0.7))
Original Graph
ggplot(top10, aes(x = name2,
y = netWorth,
fill = year2)) +
geom_bar(stat = "identity",
position = position_dodge(width = 0.9)) +
geom_text(aes(label = scales::comma(netWorth)),
angle = 90,
position = position_dodge(width = 0.9),
hjust = 1.2,
color = "white") +
theme_fivethirtyeight() +
scale_y_continuous(labels = scales::comma) +
xlab("Name") +
ylab("Net Worth in Million $") +
guides(fill = guide_legend(title = NULL)) +
ggtitle("Net Worth in Million $")