Exploring conditional missings with ggplot

Dealing With Missing Data in R

Nicholas Tierney

Statistician

What we are going to cover

How to use nabular data to explore how values change according to other values going missing
Explore visualizations:
- densities
- box plots
- different methods of splitting the visualization

ggplot(airquality,
       aes(x = Temp)) + 
  geom_density()

airquality %>%
  bind_shadow() %>%
  ggplot(aes(x = Temp,
             color = Ozone_NA)) + 
  geom_density()

airquality %>%
  bind_shadow() %>%
  ggplot(aes(x = Ozone_NA,
             y = Temp)) + 
  geom_boxplot()

airquality %>%
  bind_shadow() %>%
  ggplot(aes(x = Temp)) + 
  geom_density() + 
  facet_wrap(~Ozone_NA)

airquality %>%
  bind_shadow() %>%
  ggplot(aes(x = Temp,
             y = Wind)) + 
  geom_point() +
  facet_wrap(~ Ozone_NA)

airquality %>%
  bind_shadow() %>%
  ggplot(aes(x = Temp,
             y = Wind,
             color = Ozone_NA)) + 
  geom_point()

airquality %>%
  bind_shadow() %>%
  ggplot(aes(x = Temp,
             color = Ozone_NA)) + 
  geom_density()  +
  facet_wrap(~ Solar.R_NA)

Dealing With Missing Data in R