Otago SWC 20-10-2017 R session 2

session2.R

library(tidyverse)

gapminder <- read.csv('data/gapminder-FiveYearData.csv', header = TRUE, stringsAsFactors = FALSE)

str(gapminder)

dim(gapminder)
class(gapminder)

head(gapminder)
tail(gapminder)

gapminder %>% head()

# select country column
gapminder %>% select(country) %>% head()

# select country and year
gapminder %>% select(country, year) %>% head()

# remove country
gapminder %>% select(-country) %>% head()

# filtering rows
gapminder %>%  filter(year > 1990) %>% head()

# filtering rows on multiple conditions
# & AND
# | OR
gapminder %>%  filter(year > 1990 & lifeExp > 50) %>% head()

gapminder %>%  filter(year > 1990 & lifeExp > 50) %>%  select(country, year, lifeExp) %>% head()

# base R version
gapminder[ gapminder$year > 1990 & gapminder$lifeExp > 50, c("country", "year","lifeExp")]


# challenge
# rows with year 1990 (==) and country is Albania and just those columns (country, year)

ablania1990above <- gapminder %>% filter(year > 1990 & country == "Albania") %>% select(year, country)

# ordering rows
gapminder %>% arrange(year) %>% head()

gapminder %>% arrange(desc(year)) %>% head()

gapminder %>% arrange(desc(year), continent) %>% head()

# creating new column
gapminder %>% mutate(gdp = pop * gdpPercap) %>% head()

gapminder %>% mutate(year1990  = ifelse(year == 1990, TRUE, FALSE)) %>% head()


# groupings
gapminder %>% str()
gapminder %>% group_by(continent) %>% str()

# tally
gapminder %>% tally()

gapminder %>% group_by(continent) %>% tally()

# summarise
gapminder %>% summarise(mean_pop = mean(pop))

gapminder %>% group_by(continent) %>% summarise(mean_pop = mean(pop))

gapminder %>% group_by(continent) %>% summarise(mean_pop = mean(pop), sd_pop = sd(pop))

# challenge
# mean pop, grouped by country with largest mean as first row
gapminder %>% group_by(country) %>% summarise( mean(pop)) %>% arrange(desc(mean_pop))

# extra:
# mean gdp by continent by year
gapminder %>% mutate(gdp = gdpPercap * pop) %>% group_by(continent, year) %>% summarise(mean_gdp = mean(gdp))

# plotting
ggplot(data = gapminder, aes(x = year, y = lifeExp)) 

ggplot(gapminder, aes(x = year, y = lifeExp)) + geom_point()

ggplot(gapminder, aes(x = year, y = lifeExp)) + geom_line()

# lines, groupings, colour
ggplot(gapminder, aes(x = year, y = lifeExp, by = country, colour = continent)) + geom_line()

# layering
ggplot(gapminder, aes(x = year, y = lifeExp, by = country, colour = continent)) + geom_line() + geom_point()

#specific aes for a layer
ggplot(gapminder, aes(x = year, y = lifeExp, by = country)) + geom_line(aes(colour = continent)) + geom_point()

# tranform axis
ggplot(gapminder, aes(x = gdpPercap, y = lifeExp )) + geom_point()
ggplot(gapminder, aes(x = gdpPercap, y = lifeExp )) + geom_point() + scale_x_log10()

# statistics
last_plot() + geom_smooth(method = 'lm')

my_plot <- ggplot(gapminder, aes(x = gdpPercap, y = lifeExp )) + geom_point() + scale_x_log10()
my_plot
my_plot + geom_smooth(method = 'lm')

# facetting
az.countries <- gapminder %>% mutate( starts.with = substr(country,start = 1, stop =1)) %>% filter(starts.with %in% c('A','Z'))

ggplot(az.countries, aes(x = year, y = lifeExp, colour = continent)) + geom_line() + facet_wrap(~ country )

# themes
ggplot(az.countries, aes(x = year, y = lifeExp, colour = continent)) + geom_line() + facet_wrap(~ country ) + labs(x = 'Year', y = "Life Expectancy", title = "Figure 1", colour = "Continent") + theme(axis.text.x =element_blank(), axis.ticks.x = element_blank())

write.table(x = az.countries, file = 'data/az_countries.csv', col.names = TRUE, row.names = FALSE, quote = FALSE, sep =',')

my_plot <- ggplot(az.countries, aes(x = year, y = lifeExp, colour = continent)) + geom_line() + facet_wrap(~ country ) + labs(x = 'Year', y = "Life Expectancy", title = "Figure 1", colour = "Continent") + theme(axis.text.x =element_blank(), axis.ticks.x = element_blank())

ggsave(plot = my_plot, filename = 'figure1.png' )

# Functions

my_function <- function(){
  print("my first function")
}

arg_function <- function(word, num){
  print(word)
  print(num * 5)
}


my_plot <- function(data){
  plot(ggplot(data, aes(x = year, y = gdpPercap)) + geom_point())
}

for(cou in unique(az.countries$country)){
  my_plot(data = az.countries %>% filter(country == cou))
}