Data Visualization (GSO, Part 1)
Nguyen Chi Dung
#-------------------------------------
# Scatter Plot + Regression Line
#-------------------------------------
library(tidyverse)
# Load dữ liệu:
library(gapminder)
data("gapminder")
# Tìm hiểu về ý nghĩa của các biến số của bộ dữ liệu:
?gapminder
# Scatter Plot cơ bản:
gapminder %>%
ggplot(aes(gdpPercap, lifeExp)) +
geom_point()
# Làm mờ các điểm + tô màu theo ý muốn:
library(tidyverse)
gapminder %>%
ggplot(aes(gdpPercap, lifeExp)) +
geom_point(alpha = 0.2, color = "purple")
# Nhấn mạnh đến một số Outliers"
gapminder %>%
ggplot(aes(gdpPercap, lifeExp)) +
geom_point(alpha = 0.2, color = "purple") +
geom_point(data = gapminder %>% filter(gdpPercap > 60000),
aes(gdpPercap, lifeExp), color = "red", size = 2) ->> p1
p1
# Có thể đổi theme nếu muốn:
p1 + theme_linedraw()
p1 + theme_light()
# Hoặc cố định theme:
theme_set(theme_minimal())
p1
# Hiệu chỉnh trục X:
p2 <- p1 +
scale_x_log10()
# Hiệu chỉnh tên trục + tiêu đề:
gapminder %>%
ggplot(aes(gdpPercap, lifeExp)) +
geom_point(alpha = 0.2, color = "purple") +
scale_x_log10() +
scale_y_continuous(breaks = seq(20, 90, 10)) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/")
# Hiệu chỉnh tiếp cho trục x:
library(scales)
gapminder %>%
ggplot(aes(gdpPercap, lifeExp)) +
geom_point(alpha = 0.2, color = "purple") +
scale_x_log10(labels = scales::dollar) +
scale_y_continuous(breaks = seq(20, 90, 10)) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") ->> g3
g3
# Thêm đường hồi quy:
g3 + geom_smooth(method = "lm")
g3 + geom_smooth(method = "lm", color = "blue", fill = "blue", alpha = 0.1)
g3 + geom_smooth(method = "lm", color = "orange", se = FALSE)
# Nhấn mạnh đến Việt Nam năm 2007 chẳng hạn:
g3 +
geom_smooth(method = "lm", color = "orange", se = FALSE) +
geom_point(data = gapminder %>% filter(year == 2007 & country == "Vietnam"),
aes(gdpPercap, lifeExp), color = "red", size = 3) +
geom_text(data = gapminder %>% filter(year == 2007 & country == "Vietnam"),
aes(label = country))
# Hiệu chỉnh thêm nữa:
library(ggrepel)
g3 +
geom_smooth(method = "lm", color = "orange", se = FALSE) +
geom_point(data = gapminder %>% filter(year == 2007 & country == "Vietnam"),
aes(gdpPercap, lifeExp), color = "red", size = 3) +
geom_text_repel(data = gapminder %>% filter(year == 2007 & country == "Vietnam"),
aes(label = country), force = 19)
# Câu hỏi mở rộng: Hiển thị thêm Ấn Độ, Trung Quốc, Thái Lan và Malaysia trong
# năm 2007 nhằm so sánh các quốc gia này với Việt Nam về tuổi thọ binh quân bằng
# cách nào?
# Biểu đồ phân tán với màu sắc riêng biệt ứng với từng châu lục:
gapminder %>%
ggplot(aes(gdpPercap, lifeExp, color = continent)) +
geom_point(alpha = 0.4)
# Hoặc bỏ hiển thị legend:
gapminder %>%
ggplot(aes(gdpPercap, lifeExp, color = continent)) +
geom_point(alpha = 0.4, show.legend = FALSE)
# Hoặc điều chỉnh vị trí của legend (kiểu 1):
gapminder %>%
ggplot(aes(gdpPercap, lifeExp, color = continent)) +
geom_point(alpha = 0.4) +
theme(legend.position = c(0.9, 0.3))
# Điều chỉnh vị trí legend (kiểu 2):
gapminder %>%
ggplot(aes(gdpPercap, lifeExp, color = continent)) +
geom_point(alpha = 0.4) +
theme(legend.position = "top")
# Biểu diễn kiểu khác:
gapminder %>%
ggplot(aes(gdpPercap, lifeExp, color = continent)) +
geom_point(alpha = 0.4) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/")
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.4) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/")
# Thay đổi vị trí của legend (kiểu 1):
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.4) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") +
theme(legend.position = "top")
# Thay đổi vị trí củalLegend (kiểu 2):
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.4) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") +
theme(legend.position = c(0.85, 0.25))
# Scatter Plot trên 5 panel riêng biệt:
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.4, show.legend = FALSE) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") +
facet_wrap(~ Continent)
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.4, show.legend = FALSE) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") +
facet_wrap(~ Continent, scales = "free")
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.4, show.legend = FALSE) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") +
facet_wrap(~ Continent, nrow = 1, ncol = 5)
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.3, show.legend = FALSE) +
geom_smooth(method = "lm", color = "orange", fill = "orange", alpha = 0.2) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") +
facet_wrap(~ Continent, scales = "free")
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.3, show.legend = FALSE) +
geom_smooth(method = "lm", color = "orange", fill = "orange", alpha = 0.2) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") +
facet_wrap(~ Continent)
# Tô màu theo ý muốn:
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.3, show.legend = FALSE) +
geom_smooth(method = "lm", color = "green", fill = "green", alpha = 0.2) +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") +
facet_wrap(~ Continent) +
scale_color_manual(values = c("red", "blue", "purple", "#E69F00", "black"))
gapminder %>%
rename(Continent = continent) %>%
ggplot(aes(gdpPercap, lifeExp, color = Continent)) +
geom_point(alpha = 0.3, show.legend = FALSE) +
geom_smooth(method = "lm") +
scale_x_log10(labels = scales::dollar) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/") +
theme(legend.position = c(0.85, 0.25))
# Đường hồi quy cho Americas nhưng vẫn để các thằng kia (không phải Americas) làm nền:
gapminder %>%
ggplot(aes(gdpPercap, lifeExp)) +
scale_x_log10(labels = scales::dollar) +
geom_point(data = gapminder %>% filter(continent != "Americas"), alpha = 0.1) +
geom_point(data = gapminder %>% filter(continent == "Americas"), color = "purple") +
geom_smooth(data = gapminder %>% filter(continent == "Americas"),
method = "lm", color = "orange", fill = "orange", alpha = 0.1) +
labs(x = "GDP per Capita",
y = "Life Expectancy",
title = "The Relationship Between GDP and Life Expectancy",
subtitle = "Note: Life Expectancy at birth, in years",
caption = "Data Source: http://www.gapminder.org/data/")
#-------------------------------------------------------------------------------
# Histogram / Density Plot
# Tham khảo: http://www.pewsocialtrends.org/2016/05/11/americas-shrinking-middle-class-a-close-look-at-changes-within-metropolitan-areas/
#-------------------------------------------------------------------------------
# Đọc dữ liệu:
library(readxl)
income <- read_excel("D:/GSO_R_Course/data_for_visualization/america_metro.xlsx",
sheet = 3, skip = 7)
# Bỏ đi cột và dòng không cần thiết:
income <- income %>%
select(-X__2) %>%
slice(-1)
# Đổi lại tên cho cột biến:
names(income) <- c("Metro", "All_99", "Lower_99", "Middle_99", "Upper_99",
"All_14", "Lower_14", "Middle_14", "Upper_14")
# Xem qua dữ liệu:
sapply(income, class)
## Metro All_99 Lower_99 Middle_99 Upper_99 All_14
## "character" "numeric" "numeric" "numeric" "numeric" "numeric"
## Lower_14 Middle_14 Upper_14
## "numeric" "numeric" "numeric"
# Ba kiểu Histogram:
p1 <- income %>%
ggplot(aes(x = All_14)) +
geom_histogram(binwidth = 1000) +
labs(title = "Bin Width = 1000")
p2 <- income %>%
ggplot(aes(x = All_14)) +
geom_histogram(binwidth = 5000, color = "red", fill = "blue", alpha = 0.3) +
labs(title = "Bin Width = 5000")
p3 <- income %>%
ggplot(aes(x = All_14)) +
geom_histogram(binwidth = 10000, color = "grey30", fill = "white") +
labs(title = "Bin Width = 10000")
library(gridExtra)
grid.arrange(p1, p2, p3, nrow = 1, ncol = 3)
# Density Plot:
income %>%
ggplot(aes(x = All_14)) +
geom_density()
income %>%
ggplot(aes(x = All_14)) +
geom_density(color = "grey40", fill = "grey80", size = 1.2)
# Biểu diễn đồng thời cả Histogram và Density:
income %>%
ggplot(aes(x = All_14)) +
geom_density(color = "red", fill = "red", alpha = 0.2) +
geom_histogram(aes(y = ..density..), binwidth = 2000,
fill = "blue", color = "blue", alpha = 0.2)
# So sánh các nhóm. Trước hết chuyển dữ liệu về long form:
compare <- income %>%
select(Metro, All_99, All_14) %>%
gather(Year, Income, -Metro)
# Tham khảo thêm cách chọn màu ở http://colorbrewer2.org/:
compare %>%
ggplot(aes(x = Income, fill = Year)) +
geom_density(alpha = 0.4) +
scale_fill_manual(values = c("#1b9e77", "#7570b3"))
compare %>%
ggplot(aes(x = Income)) +
geom_histogram(binwidth = 2000, color = "grey30", fill = "white") +
facet_grid(Year ~ .)
class_comparison <- income %>%
select(Metro, Lower_99:Upper_99, Lower_14:Upper_14) %>%
gather(Class, Income, -Metro) %>%
separate(Class, into = c("Class", "Year")) %>%
mutate(Year = case_when(Year == "99" ~ "1999",
Year != "99" ~ "2014"))
class_comparison %>%
ggplot(aes(x = Income, fill = Class, color = Class)) +
geom_histogram(alpha = 0.2) +
facet_wrap(~ Class, scales = "free_x")
class_comparison %>%
ggplot(aes(x = Income, fill = Class, color = Class)) +
geom_histogram(alpha = 0.2) +
facet_wrap(Year ~ Class, scales = "free_x")
class_comparison %>%
ggplot(aes(x = Income, fill = Class, color = Class)) +
geom_histogram(alpha = 0.2) +
facet_wrap(Year ~ Class, scales = "free")
class_comparison %>%
ggplot(aes(x = Income, fill = Year, color = Year)) +
geom_histogram(alpha = 0.2) +
facet_wrap(Year ~ Class, scales = "free")
# Thu nhập trung bình của từng nhóm theo năm:
class_mean <- class_comparison %>%
group_by(Class, Year) %>%
summarise(Mean = mean(Income)) %>%
ungroup()
class_comparison %>%
ggplot(aes(x = Income, fill = Class, color = Class)) +
geom_histogram(alpha = 0.2, show.legend = FALSE) +
facet_wrap(Year ~ Class, scales = "free") +
geom_vline(data = class_mean, aes(xintercept = Mean), linetype = "dashed") +
ylab("Frequency") +
xlab("Median Household Income (thousands)") +
labs(title = "Median Household Income by Income Tier Across U.S. Metropolitan Areas",
subtitle = "Average median income across 229 metros decreased from $67,863 in 1999 to $62,662 in 2014, representing an 8% loss in \nincome. The lower income class experienced the largest impact with a 11% decrease while the middle and upper class median \nhousehold income decreased by 6% and 8% respectively.",
caption = "Source: Pew Research Center analysis of the \n2000 decennial census and 2014 American \nCommunity Survey (IPUMS)")
# Cải tiến:
p <- class_comparison %>%
ggplot(aes(x = Income / 1000, fill = Class, color = Class)) +
geom_histogram(alpha = 0.2, show.legend = FALSE) +
facet_wrap(Year ~ Class, scales = "free") +
geom_vline(data = class_mean, aes(xintercept = Mean / 1000), linetype = "dashed") +
scale_x_continuous(labels = scales::dollar) +
scale_y_continuous(limits = c(0, 58), expand = c(0, 0)) +
labs(x = "Median Household Income (thousands)", y = "Frequency",
title = "Median Household Income by Income Tier Across U.S. Metropolitan Areas",
subtitle = "Average median income across 229 metros decreased from $67,863 in 1999 to $62,662 in 2014, representing an 8% loss in \nincome. The lower income class experienced the largest impact with a 11% decrease while the middle and upper class median \nhousehold income decreased by 6% and 8% respectively.",
caption = "Source: Pew Research Center analysis of the \n2000 decennial census and 2014 American \nCommunity Survey (IPUMS)")
p
# Cải tiến hơn nữa:
class_mean <- class_mean %>%
mutate(Label = paste0("$", prettyNum(round(Mean, 0), big.mark = ",")))
p +
geom_text(data = class_mean,
aes(x = Mean / 1000, y = 52, id = Class, label = Label),
size = 3, hjust = -.1, show.legend = FALSE, color = "black")
#------------------------
# Box Plot
#------------------------
# Vẽ đơn giản:
iris %>%
ggplot(aes(Species, Sepal.Length)) +
geom_boxplot()
# Tô màu đỏ, chẳng hạn, nếu muốn nhấn mạnh đến outliers:
iris %>%
ggplot(aes(Species, Sepal.Length)) +
geom_boxplot(outlier.color = "red")
# Hiển thị thêm mean nếu muốn:
iris %>%
ggplot(aes(Species, Sepal.Length)) +
geom_boxplot(outlier.color = "red") +
stat_summary(fun.y = mean, colour = "blue", geom = "point")
# Boxplot của cả 4 biến số cho 3 loài hoa diên vĩ + đầy đủ chỉ dẫn:
iris %>%
gather(Variable, Value, -Species) %>%
ggplot(aes(Species, Value, fill = Species, color = Species)) +
geom_boxplot(show.legend = FALSE, alpha = 0.4) +
facet_wrap(~ Variable, scales = "free") +
labs(x = NULL,
y = NULL,
title = "An Example of Boxplot",
caption = "Data Source: Iris data set by R. Fisher (1936)")
# Hoặc một kiểu khác:
iris %>%
gather(Variable, Value, -Species) %>%
ggplot(aes(Species, Value, fill = Species, color = Species)) +
geom_boxplot(show.legend = FALSE, alpha = 0.4) +
facet_wrap(~ Variable, scales = "free") +
theme_bw() +
labs(x = NULL,
y = NULL,
title = "An Example of Boxplot",
caption = "Data Source: Iris data set by R. Fisher (1936)")
# So sánh với Histogram và Density:
iris %>%
gather(Variable, Value, -Species) %>%
ggplot(aes(Value, fill = Species, color = Species)) +
geom_histogram(alpha = 0.3) +
facet_wrap(~ Variable) +
theme_bw() +
theme(legend.position = "top") +
labs(x = NULL,
y = NULL,
title = "An Example of Histogram",
caption = "Data Source: Iris data set by R. Fisher (1936)")
iris %>%
gather(Variable, Value, -Species) %>%
ggplot(aes(Value, fill = Species, color = Species)) +
geom_density(alpha = 0.3, show.legend = FALSE) +
facet_wrap(~ Variable, scales = "free") +
theme_bw() +
labs(x = NULL,
y = NULL,
title = "An Example of Density Plot",
caption = "Data Source: Iris data set by R. Fisher (1936)")
iris %>%
gather(Variable, Value, -Species) %>%
ggplot(aes(Value, fill = Species, color = Species)) +
geom_density(alpha = 0.3, show.legend = FALSE) +
facet_wrap(~ Variable) +
theme_bw() +
labs(x = NULL,
y = NULL,
title = "An Example of Density Plot",
caption = "Data Source: Iris data set by R. Fisher (1936)")
