---
title: Visualising Missing Data
author: Hugo Soubrier
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Visualising Missing Data}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

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

```{r setup}
library(dplyr)
library(ggplot2)
library(epivis)
```

```{r load-data}
df <- epivis::moissala_measles 

glimpse(df)
```

By default `plot_miss_vis()` will plot a graph of the missing values in each of the observations and variables of the dataframe. It will present the proportions of missing values for a single variables in the label of the y-axis, and the overall proportions of missing value across the dataframe in the legend.

```{r}
plot_miss_vis(
  df
)
```


## Facet the graph 

The missing values plot can be facetted by a group variable using the `facet = "variable_name"` argument

```{r}

df |> 
  filter(site %in% c("Moïssala Hospital", "Bedaya Hospital")) |> 
  plot_miss_vis(
    facet = "site"
  )

```


## Further customisation

Colors of the graph can be changed using the `color_vec` argument. It takes a length 2 vector of HEX code, with the first color for Missing values. `y_axis_text_size` allows manual specification of the y-axis text size, which can be hard to read when exploring many variables.

```{r}

df |> 
  filter(site %in% c("Moïssala Hospital", "Bedaya Hospital")) |> 
  plot_miss_vis(
    facet = "site", 
    y_axis_text_size = 6, 
    col_vec = c("seagreen", "lightblue")
  )

```