BIOS 591R - 2019 - Lecture 8: Graphics I: base graphics

Ziyi Li <ziyi.li@emory.edu>

R graphics can be confusing because there are no less than 4 different systems. Let's list them out here and talk about which one(s) to use.

Low-Level Capability

High-Level Capability

Oldest and most commonly used
Uses a “pen-on-paper” model. You can only draw on top of the object. Cannot erase, modify, or delete what has already been drawn. Mimicing the thinking process.
Has both high and low level plotting routines (unique to Base)
Base graphics are fast.
Lots of documentation and “google” support

Developed in 2000 by Paul Murrell
Provides a rich set of graphics primitives
Uses a system of objects and view ports to make complex objects easier.
You will almost never use this directly unless you want to do in-depth programming

Developed by Deepayan Sarkar to implement the trellis graphics system described in “Visualizing Data” by Cleveland.
Independent from the BASE graphics. Include its own functions such as xyplot, bwplot, levelplot
Easy to create conditioned plots with automatic creation of axes, legends, and other annotations
Usually considered to be an improvement over Base graphics.

library(lattice)
xyplot(mpg~wt | factor(am,labels=c("Auto","Manual")), data=mtcars)

plot of chunk unnamed-chunk-1

Developed starting in 2005 by Hadley Wickham
ggplot2 is an implementation of Leland Wilkinson's Grammar of Graphics–a general scheme for data visualization which breaks up graph into semantic components such as scales and layers.
ggplot2 can serve as a replacement for the base graphics in R and contains a number of defaults for web and print display of common scales.
Is said to be much slower than Base graphics but this isn't a major thing (in my opinion)
Part of Tidyverse

BASE Graphics