Created
August 24, 2017 15:08
-
-
Save psobczyk/49d6cbf855a9407780f5826daaa6b153 to your computer and use it in GitHub Desktop.
Code for visualizations in blog post http://szychtawdanych.pl/?p=1704
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
library(rvest) | |
library(ggplot2) | |
library(ggthemes) | |
library(highcharter) | |
avg_temp <- read_html('https://en.wikipedia.org/wiki/List_of_countries_by_average_yearly_temperature') | |
avg_temp <- html_table(avg_temp, fill = T)[[2]] | |
gdp <- read_html('https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(PPP)_per_capita') | |
gdp_tables <- html_table(gdp, fill = T) | |
gdp <- gdp_tables[[3]] | |
hdi <- read_html('https://en.wikipedia.org/wiki/List_of_countries_by_inequality-adjusted_HDI') | |
hdi_tables <- html_table(hdi, fill = T) | |
hdi <- bind_rows(hdi_tables[[3]], hdi_tables[[4]]) | |
names(avg_temp)[3] <- "foo" | |
avg_temp <- avg_temp %>% | |
filter(is.na(foo)) %>% | |
select(country = Country, | |
temperature = "Average yearly temperature (1961-1990, Celsius)") %>% | |
mutate(temperature = as.numeric(temperature)) %>% | |
arrange(desc(temperature)) %>% | |
mutate(rank = row_number()) | |
gdp <- gdp %>% | |
select(rank = Rank, country = "Country/Territory", | |
gdp_ppp = "Int$") %>% | |
mutate(country = if_else(country == "China[n 1]", "China", country)) %>% | |
rowwise %>% | |
mutate(gdp_ppp = as.numeric(gsub("[[:space:]]*,[[:space:]]*", "", gdp_ppp))) | |
plot_data <- inner_join(gdp, avg_temp, by = "country") %>% | |
inner_join(hdi, by = c("country"="Country")) | |
breaks_gdp <- c(0, 1e4, 2e4, 4e4,8e4, 16e4) | |
plot_data <- plot_data %>% | |
mutate(gdp_label = cut(gdp_ppp, breaks_gdp, labels = c("< 10 tys.$", "10 - 20 tys.$", | |
"20 - 40 tys.$", | |
"40 - 80 tys.$", | |
"80 - 120 tys.$"))) | |
ggplot(plot_data, aes(x = temperature, y = IHDI, color = gdp_label)) + | |
geom_point() + | |
theme_fivethirtyeight(base_size = 20, base_family = "Helvetica Neue Light") + | |
labs(title = "Wpływ temperatury na rozwój państw", | |
x = "Przeciętna temperatura roczna", | |
y = "Human Development Index") + | |
guides(color = guide_legend(title = "PKB per capita", override.aes = aes(size = 6))) + | |
theme(axis.title = element_text(inherit.blank = F), | |
plot.title = element_text(hjust = 0.5), | |
panel.background = element_rect(fill = "white"), | |
plot.background = element_rect(fill = "white"), | |
legend.background = element_rect(fill = "white"), | |
legend.key = element_rect(fill = "white")) + | |
ggplot2::annotate("text", x = 0, y = 0.3, label = "Korelacja: -0.7", | |
hjust=0, vjust=0, col="black", cex=6, | |
fontface = "bold", alpha = 0.5) | |
hchart(plot_data, "scatter", hcaes(x = temperature, y = IHDI, | |
label = country, group = gdp_label)) %>% | |
hc_title(text = "Przeciętna temperatura i rozwój krajów", | |
align = "center", | |
style = list(useHTML = TRUE)) %>% | |
hc_xAxis(title = list(text = "Przeciętna temperatura")) %>% | |
hc_yAxis(title = list(text = "Human Development Index")) %>% | |
hc_add_theme(hc_theme_538()) %>% | |
hc_chart(zoomType = "xy") %>% | |
hc_tooltip(useHTML = TRUE, | |
headerFormat = "<table>", | |
pointFormat = paste("<tr><th colspan=\"1\"><b>{point.label}</b></th></tr>", | |
"<tr><th>Przeciętna temperatura</th><td>{point.x} C</td></tr>", | |
"<tr><th>Human Development Index</th><td>{point.y} </td></tr>"), | |
footerFormat = "</table>") | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment