README.md

In alexhallam/gumballthemes:

Installation

This package isn't on CRAN, so you'll need to use the devtools package to install it.

# If you don't have devtools installed
install.packages("devtools")

devtools::install_github("alexhallam/gumballthemes")

# To include the vignette
devtools::install_github("alexhallam/gumballthemes", build_vignettes=TRUE)

Introduction

The gumballthemes package has three goals/motivations.

1. Quantile frames - for quick insights
2. Color blind friendly colors - be a decent human being, make plots everyone can interpret
3. Readable Fonts - easy reading as asthetics (not implimented yet)

Base Theme

This is a basic example which shows how to use this theme.

## basic example code
p <- mtcars %>%
  as_tibble() %>%
  ggplot(aes(y = wt, x = mpg)) +
  geom_point() +
  geom_quantileframe() +
  scale_y_quantile(mtcars$wt) +
  scale_x_quantile(mtcars$mpg, digits = 1) +
  ylab("Weight (tons)") +
  xlab("Fuel efficiency (mpg)") +
  scale_color_gumball() +
  theme_gumball() +
  guides(col = guide_legend(ncol= 6)) 

p %>% ggsave(filename = "basic_img.JPG", height = 3.5, width = 5.95)

The above plots uses all of the functions currently provided by gumballthemes:

scale_y_quantile

scale_x_quantile

theme_gumball

The data-pixel ratio is high in the above plot. The range of each of the axes are the min and max values. The thick bar represents the 25th, 50th (aka median), and 75th percentiles. The axes are essentially simple boxplots.

Base Theme With hrbrthemes for better fonts

hrbrthemes offers opinionated, typographic-centric ggplot2 themes. These can be combined with geom_quantileframe plots. The result is a plots with cleaner text.

library(extrafont)
#> Registering fonts with R
p <- mtcars %>%
  as_tibble() %>%
  ggplot(aes(y = wt, x = mpg)) +
  geom_point() +
  geom_quantileframe() +
  scale_y_quantile(mtcars$wt) +
  scale_x_quantile(mtcars$mpg, digits = 1) +
  ylab("Weight (tons)") +
  xlab("Fuel efficiency (mpg)") +
  scale_color_gumball() +
  hrbrthemes::theme_ipsum(grid = F) +
  guides(col = guide_legend(ncol= 6))

p %>% ggsave(filename = "basic_hrbrh_img.JPG", height = 3.5, width = 5.95)

Base Theme + Viridis color

The viridis package is usefull for producing themes that are color blind friendly

p <- mtcars %>%
  as_tibble() %>%
  ggplot(aes(y = wt, x = mpg, color = factor(carb))) +
  geom_point() +
  geom_quantileframe() +
  scale_y_quantile(mtcars$wt) +
  scale_x_quantile(mtcars$mpg, digits = 1) +
  ylab("Weight (tons)") +
  xlab("Fuel efficiency (mpg)") +
  scale_color_gumball() +
  theme_gumball()+
  guides(col = guide_legend(ncol= 6)) 

p %>% ggsave(filename = "basic_color_img.JPG", height = 3.5, width = 5.95)

Time Series

There may also be opportunity or a quntile frame on the y-axis only on time-series plots.

The below example requiers an API key from FRED.

#library(fredr)
#fredr_set_key("abcdefghijklmnopqrstuvwxyz123456")
unrate <- fredr_series_observations(
  series_id = "UNRATE",
  observation_start = as.Date("1990-01-01"),
  observation_end = as.Date("2018-04-01"),
  frequency = "q",
  units = "chg"
)

unrate %>% glimpse()
#> Observations: 114
#> Variables: 3
#> $ date      <date> 1990-01-01, 1990-04-01, 1990-07-01, 1990-10-01, 199...
#> $ series_id <chr> "UNRATE", "UNRATE", "UNRATE", "UNRATE", "UNRATE", "U...
#> $ value     <dbl> -0.06666667, 0.03333333, 0.36666667, 0.43333333, 0.4...

p <- unrate %>% 
  ggplot(aes(y = value, x = date)) +
  geom_line() +
  geom_quantileframe(sides = "l") +
  scale_y_quantile(unrate$value) + 
  scale_x_date(date_breaks = "5 year", date_labels = "%Y") +
  ylab("Change In UNRATE") +
  xlab("") +
  annotate("text", y = 1.0, x = as.Date("1997-01-01"),
           label = "Change in UNRATE\nis normally between\n0.07% and -.20% with a\nmedian of -0.07") +
  guides(col = guide_legend(ncol= 6)) +
  scale_color_gumball() +
  theme_gumball()

p %>% ggsave(filename = "time_series.JPG", height = 3.5, width = 5.95)

Related Work

hrbrthemes

ggthemes

ggplot2

alexhallam/gumballthemes documentation built on May 15, 2019, 11:49 a.m.