---
title: "Inequality Measurement in DCEA"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Inequality Measurement in DCEA}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include = FALSE}
knitr::opts_chunk$set(collapse = TRUE, comment = "#>")
library(dceasimR)
```

## Overview

`dceasimR` provides five inequality measures commonly used in DCEA:
SII, RII, concentration index, Atkinson index, and Gini coefficient.

```{r data}
df <- tibble::tibble(
  group      = 1:5,
  mean_hale  = c(52.1, 56.3, 59.8, 63.2, 66.8),
  pop_share  = rep(0.2, 5)
)
```

## Slope Index of Inequality (SII)

The SII estimates the absolute health difference from the most to the least
deprived using a weighted regression on ridit scores.

```{r sii}
calc_sii(df, "mean_hale", "group", "pop_share")
```

A positive SII means better health in more advantaged groups.

## Relative Index of Inequality (RII)

The RII expresses the SII relative to mean health, facilitating comparisons
across populations and time.

```{r rii}
calc_rii(df, "mean_hale", "group", "pop_share")
```

## Concentration Index

```{r ci}
calc_concentration_index(df, "mean_hale", "group", "pop_share",
                          type = "standard")
```

## Atkinson Index

```{r atkinson}
calc_atkinson_index(df$mean_hale, df$pop_share, epsilon = 1)
```

## Gini Coefficient

```{r gini}
calc_gini(df$mean_hale, df$pop_share)
```

## All indices at once

```{r all}
calc_all_inequality_indices(df, "mean_hale", "group", "pop_share",
                             epsilon_values = c(0.5, 1, 2))
```

## Lorenz curves

```{r lorenz-data}
ld <- compute_lorenz_data(df$mean_hale, df$pop_share, "England 2019")
```

```{r lorenz-plot, fig.width = 5, fig.height = 4}
library(ggplot2)
ggplot(ld, aes(cum_pop, cum_health)) +
  geom_line(colour = "steelblue", linewidth = 1) +
  geom_abline(linetype = "dashed") +
  labs(x = "Cumulative population", y = "Cumulative health",
       title = "Lorenz Curve") +
  theme_minimal()
```