# Inequality Key Figures - replication with R and `lissyrtools`

library(lissyrtools)
library(tidyverse)


# 0) Load Data and Prepare it 
data <- lissyuse(
  data = "lu",   # Examples: `data = c("de", "it")`; `data = c("it20", "it16")`; Run: get_countries_lis() to check other iso2 codes, or get_years_lis(c("it", "es")) to check which years are available.
  vars = c("dhi", "nhhmem65", "nhhmem17","sex", "relation", "hpartner")
) %>%
  map(~ .x %>% 
        filter(!is.na(dhi)) %>%
        filter(relation == 1000) %>%
        mutate(
          new_wgt = hwgt*nhhmem, 
          new_chld_wgt = hwgt * nhhmem17,
          new_eldr_wgt = hwgt * nhhmem65 
        )
  ) %>%
  apply_iqr_top_bottom_coding("dhi", "hwgt", type = "type_4") %>% 
  apply_sqrt_equivalisation("dhi")


medians <- map(
  run_weighted_percentiles(data, "dhi", "new_wgt", type = "type_4"), 
  "50%"
)




# Distribution of Children by Income Group (h-level data)

for (i in c(0.5, 0.75, 1.5)) {
  
  assign("poverty_line", map(medians  , ~ .x  * i )) # multiply the median by (0.5, 0.75, or 1.5), thus defining a poverty line
  
  data_chld <- map2(
    data, 
    poverty_line,
    ~ .x %>%
      mutate(
        line = .y,
        poor = if_else(dhi < line, 100, 0)
      )
  )
  
  assign(paste0("perc_poor_", i),
         run_weighted_mean(data_chld, "poor", "new_chld_wgt"))
  
}

dist_child_75_50 <- map2(perc_poor_0.75, perc_poor_0.5, ~ .x - .y)
dist_child_150_75 <- map2(perc_poor_1.5, perc_poor_0.75, ~ .x - .y)
dist_child_above_150 <- map(perc_poor_1.5, ~ (100 - .x))





# Collect them into a list
all_stats <- list(
  d5075 = dist_child_75_50,
  d75150 = dist_child_150_75,
  d150 = dist_child_above_150
)

results <- map(all_stats,  ~ structure_to_plot(.x, print_columns = FALSE)) 

final <- imap(results , ~ .x %>% mutate(short_ikf = .y)) %>% 
  bind_rows() %>% 
  arrange(short_ikf, dname) %>%  
  mutate(
    indicator = case_when(
      short_ikf == "d5075"      ~ "Distribution of Children by Income Group (50 75%)",
      short_ikf == "d75150"     ~ "Distribution of Children by Income Group (75 150%)",
      short_ikf == "d150"       ~ "Distribution of Children by Income Group (above 150%)"
    )
  ) %>%
  rename(value_ikf_R = value)

write.csv(final)