vignettes/Plotting.md

In wuergh/hedgehog: R functions for personal use

title: "Plotting" author: "Erle Holgersen" date: "2018-03-17" output: html_document: keep_md: true vignette: > %\VignetteIndexEntry{Vignette Title} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8}

For visualization, I focus on writing functions for plots I use frequently and that require many lines of base R. At present, the package support two plots, km.plot and stripboxplot.

Kaplan-Meier Plots – km.plot

Kaplan-Meier plots are used for visualizing censored data. They are most commonly used in medical studies for showing how survival differs between treatment groups. Building a KM plot in R is straightforward, but requires several function calls. First you call the survfit function from the survival package, and then call plot on the resulting object.

To speed up this process, I wrote my own wrapper function km.plot.

library(hedgehog);
library(survival);

# using example from survival package
km.plot(
    Surv(futime, death) ~ sex, 
    mgus
    );

km.plot(
    Surv(futime, death) ~ sex, 
    mgus,
    show.risk.table = TRUE
    );

Box plots – stipboxplot

Standard box plots hide the sample size in each group. One way of getting around this is to add a background stripplot that shows each point in the dataset.

test.data <- data.frame(
    x = sample(letters[1:3], 100, replace = TRUE),
    y = rnorm(100)
    );

stripboxplot(y ~ x, test.data);

By default, all background points are grey. We can change this by using the points.col argument. It is passed directly to stripchart, and thus follows the stripchart convention of assigning colours by grouping variable.

stripboxplot(
    y ~ x, 
    test.data,
    points.col = c('darkgreen', 'orange', 'firebrick')
    );

There are a number of arguments for tweaking the plot appearance. To change the category labels, use the group.names argument.

stripboxplot(
    y ~ x, 
    test.data,
    points.col = c('darkgreen', 'orange', 'firebrick'),
    group.names = c('Y', 'A', 'Y')
    );

Colour palettes

The function get.colour.palette contains a few pre-made colour palettes. To quickly look at what colours are included in a palette, run show.colour.palette.

show.colour.palette( get.colour.palette(7) );

The visualization function also works with other colour palettes.

show.colour.palette( colours(7) );

wuergh/hedgehog documentation built on June 2, 2019, 7:55 p.m.