In jcpernias/ec1047: Code and data for the EC1047 course

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(tibble)
library(lubridate, warn.conflicts = FALSE)
library(ggplot2)
library(dplyr, warn.conflicts = FALSE)
library(tidyr)
library(kableExtra, warn.conflicts = FALSE)
library(ec1047)

# Suppress summarise info
options(dplyr.summarise.inform = FALSE)

Variables

str(ecv)

Disposable income

A relatively small number of observations have negative or zero values of disposable income:

ecv %>% filter(ydisp_hh <= 0.0) %>% 
  group_by(year) %>% summarise(n = n())

Drop observations with zero or negative disposable income. Also drop observations from Ceuta y Melilla:

income_db <- ecv %>% 
  filter(ydisp_hh > 0.0, !(region %in% c("CEU", "MEL")))

Mean household income by year:

income_db %>% group_by(year) %>%
  summarise(y_mean = wtd_mean(ydisp_hh, weight))

Mean disposable income per person by year:

income_db %>%
  mutate(year = lubridate::make_date(year = year)) %>% 
  group_by(year) %>%
  summarise(y_mean = wtd_mean(ydisp_cu, weight * people)) %>%
  ggplot(aes(x = year, y = y_mean)) + 
  geom_line()

Mean disposable income per person by year and region:

y_mean_esp <- income_db %>% group_by(year) %>%
  summarise(y_mean = wtd_mean(ydisp_cu, weight * people)) %>%
  mutate(region = "ESP")

y_mean_reg <- income_db %>%
  mutate(region = as.character(region)) %>% 
  group_by(region, year) %>%
  summarise(y_mean = wtd_mean(ydisp_cu, weight * people)) 

bind_rows(y_mean_reg, y_mean_esp) %>%
  pivot_wider(names_from = year, values_from = y_mean) %>%
  kable(digits = 0) %>%
  kable_styling(bootstrap_options = c('striped', 'hover'), 
                full_width = FALSE)

Some graphics:

y_mean_reg %>%
  mutate(year = lubridate::make_date(year = year)) %>%
  ggplot(aes(x=year, y = y_mean)) +
  geom_line() + facet_wrap(~ region, nrow = 4)