yarrr
A Companion to the e-Book "YaRrr!: The Pirate's Guide to R"
Contains a mixture of functions and data sets referred to in the introductory e-book "YaRrr!: The Pirate's Guide to R". The latest version of the e-book is available for free at < https://www.thepiratesguidetor.com>.
YaRrr Package
The yarrr package contains a mixture of data, functions and tutorials supporting the e-book "YaRrr! The Pirate's Guide to R" (www.thepiratesguidetor.com).
To install the (stable) version from CRAN, run the following code
install.packages("yarrr")
To install the latest developer version from GitHub, run the following code
# Install developer version from GitHubinstall.packages("devtools")devtools::install_github("ndphillips/yarrr",build_vignettes = TRUE)
Once you've installed the yarrr package, you can load it and open the main package guide as follows:
library("yarrr") # load yarrryarrr.guide() # run main package guide
Here are the most important parts of the package:
pirateplot()
The pirateplot function creates a pirateplot, a transparent (both literally and figuratively) plot for displaying continuous data as a function of 1, 2, or 3 discrete variables. Unlike traditional plots, like barplots and boxplots, the pirateplot shows both raw data (jittered points), descriptive statistics (line and/or bar), and inferential statistics (95% Bayesian Highest Density Intervals or Confidence Intervals), in one plot. While the default plot shows all these elements, the user can easily customize the transparency of each element using additional arguments.
See ?pirateplot or https://CRAN.R-project.org/package=yarrr for more details
News
yarrr 0.1.5
-
Minor changes to
theme = 1inpirateplot(). Changed default value ofcap.beanstoTRUE -
Added color mixing arguments
mix.colandmix.ptopiratepal(). These allow you to mix the default palettes with a specified color (e.g.;"white") -
Added the option to specify data in
pirateplot()as a list of numeric vectors, or as a numeric dataframe or matrix without specifying a formula. Each column / element will be taken as a new group. -
New palettes in
piratepal():decision. -
Fixed bug in
sortxinpirateplot(). Sorting data by functions (e.g.sortx = "mean") should now work. -
Added
glargument topirateplot()to specify locations of gridlines (e.g.;gl = seq(0, 10, 1)) -
Added
cex.namesargument to control size of bean names (currently this was controlled bycex.lab, which now controls the size of the axis names.)
yarrr 0.1.4
-
Some minor changes to default plotting parameters that I think make the default plots look a bit nicer.
-
Added
cap.beansargument topirateplot(). Whencap.beans = TRUE, beans will be cut at the maximum and minimum values of the data. -
Added
cap.beansargument topirateplot(). Whencap.beans = TRUE, beans will be cut at the maximum and minimum values of the data. -
Added two new
inf.methodvalues:sdfor standard deviation, andsefor standard error
yarrr 0.1.2
-
Minor updates to themes. Added
theme = 3 -
You can now assign a pirateplot to a variable to return summary statistics.
yarrr 0.1.0
-
Added a
NEWS.mdfile to track changes to the package. -
Re-structured code generating colors and opacities in
pirateplot()to make future updates easier. -
Added
quant,quant.lengthandquant.widtharguments that add horizontal lines for specified quantiles to each bean (thanks @pat-s) -
Added several new arguments (e.g.;
bean.fill.colfor customising pirateplots -
Improved theme support (now under
themerather thantheme.o)
yarrr 0.0.7
pirateplot()can now handle up to 3 IVs!
Example:pirateplot(age ~ sex + headband + favorite.pirate, data = pirates).
Levels of the third IV are shown in separate plots in a grid.
Minor and Bug-fixes
- The
inf.pparameter inpirateplot()was prevously not being passed to the Bayesian HDIs, rendering all inference bands to be the default of 95% (thanks to Roman Pahl for catching this). This has now been fixed.
yarrr 0.0.6
- Added
hdi.bandargument topirateplot(). Settinghdi.band = "tight"will constrain inference bands to bean densities. - Minor changes to gridlines when specifying
gl.col.