探索性数据分析week2

2.1 the lattice绘图系统

The Lattice Plotting System

The lattice plotting system is implemented using the following packages:

• lattice: contains code for producing Trellis graphics, which are independent of the “base” graphics system; includes functions like xyplot, bwplot, levelplot

• grid: implements a different graphing system independent of the “base” system; the lattice package builds on top of grid

  • We seldom call functions from the grid package directly

• The lattice plotting system does not have a "two-phase" aspect with separate plotting and annotation like in base plotting

• All plotting/annotation is done at once with a single function call

Lattice Functions

• xyplot: this is the main function for creating scatterplots
• bwplot: box-and-whiskers plots (“boxplots”)
• histogram: histograms
• stripplot: like a boxplot but with actual points
• dotplot: plot dots on "violin strings"
• splom: scatterplot matrix; like pairs in base plotting system
• levelplot, contourplot: for plotting "image" data

Lattice Functions

Lattice functions generally take a formula for their first argument, usually of the form

xyplot(y ~ x | f * g, data)

• We use the formula notation here, hence the ~.

• On the left of the ~ is the y-axis variable, on the right is the x-axis variable

• f and g are conditioning variables — they are optional

  • the * indicates an interaction between two variables

• The second argument is the data frame or list from which the variables in the formula should be looked up

  • If no data frame or list is passed, then the parent frame is used.

• If no other arguments are passed, there are defaults that can be used.

Lattice Behavior

Lattice functions behave differently from base graphics functions in one critical way.

• Base graphics functions plot data directly to the graphics device (screen, PDF file, etc.)

• Lattice graphics functions return an object of class trellis

• The print methods for lattice functions actually do the work of plotting the data on the graphics device.

• Lattice functions return "plot objects" that can, in principle, be stored (but it’s usually better to just save the code + data).

• On the command line, trellis objects are auto-printed so that it appears the function is plotting the data

Lattice Panel Functions

• Lattice functions have a panel function which controls what happens inside each panel of the plot.

• The lattice package comes with default panel functions, but you can supply your own if you want to customize what happens in each panel

• Panel functions receive the x/y coordinates of the data points in their panel (along with any optional arguments)

其下若无panel.xyplot(x,y,...)则不会显示原函数

Many Panel Lattice Plot: Example from MAACS

• Study: Mouse Allergen and Asthma Cohort Study (MAACS)

• Study subjects: Children with asthma living in Baltimore City, many allergic to mouse allergen

• Design: Observational study, baseline home visit + every 3 months for a year.

• Question: How does indoor airborne mouse allergen vary over time and across subjects?

Ahluwalia et al., Journal of Allergy and Clinical Immunology, 2013

Summary

• Lattice plots are constructed with a single function call to a core lattice function (e.g. xyplot)

• Aspects like margins and spacing are automatically handled and defaults are usually sufficient

• The lattice system is ideal for creating conditioning plots where you examine the same kind of plot under many different conditions

• Panel functions can be specified/customized to modify what is plotted in each of the plot panels

2.2 ggplot2

What is ggplot2?

• An implementation of The Grammar of Graphics by Leland Wilkinson
• Written by Hadley Wickham (while he was a graduate student at Iowa State)
• A “third” graphics system for R (along with base and lattice)
• Available from CRAN via install.packages()
• Web site: http://ggplot2.org (better documentation)

What is ggplot2?

• Grammar of graphics represents an abstraction of graphics ideas/objects
• Think “verb”, “noun”, “adjective” for graphics
• Allows for a “theory” of graphics on which to build new graphics and graphics objects
• “Shorten the distance from mind to page”

Grammer of Graphics

“In brief, the grammar tells us that a statistical graphic is a mapping from data to aesthetic attributes (colour, shape, size) of geometric objects (points, lines, bars). The plot may also contain statistical transformations of the data and is drawn on a specific coordinate system”

• from ggplot2 book

The Basics: qplot()

• Works much like the plot function in base graphics system
• Looks for data in a data frame, similar to lattice, or in the parent environment
• Plots are made up of aesthetics 美学(size, shape, color) and geoms(几何) (points, lines)

The Basics: qplot()

• Factors are important for indicating subsets of the data (if they are to have different properties); they should be labeled
• The qplot() hides what goes on underneath, which is okay for most operations
• ggplot() is the core function and very flexible for doing things qplot() cannot do

具体应用那节，网上好像有问题，没有，所以去

https://github.com/linsidu/courses/blob/master/04_ExploratoryAnalysis/ggplot2/ggplot2_p1.md这有

What is ggplot2?

• An implementation of the Grammar of Graphics by Leland Wilkinson
• Grammar of graphics represents and abstraction of graphics ideas/objects
• Think “verb”, “noun”, “adjective” for graphics
• Allows for a “theory” of graphics on which to build new graphics and graphics objects

Basic Components of a ggplot2 Plot

• A data frame
• aesthetic mappings: how data are mapped to color, size
• geoms: geometric objects like points, lines, shapes.
• facets: for conditional plots.
• stats: statistical transformations like binning, quantiles, smoothing.
• scales: what scale an aesthetic map uses (example: male = red, female = blue).
• coordinate system

Building Plots with ggplot2

• When building plots in ggplot2 (rather than using qplot) the “artist’s palette” model may be the closest analogy
• Plots are built up in layers
  • Plot the data
  • Overlay a summary
  • Metadata and annotation

Example: BMI, PM2.5, Asthma

• Mouse Allergen and Asthma Cohort Study
• Baltimore children (age 5-17)
• Persistent asthma, exacerbation in past year
• Does BMI (normal vs. overweight) modify the relationship between PM2.5 and asthma symptoms?

Building Up in Layers

head(maacs)

  logpm25        bmicat NocturnalSympt logno2_new
1  1.5362 normal weight              1      1.299
2  1.5905 normal weight              0      1.295
3  1.5218 normal weight              0      1.304
4  1.4323 normal weight              0         NA
5  1.2762    overweight              8      1.108
6  0.7139    overweight              0      0.837

g <- ggplot(maacs, aes(logpm25, NocturnalSympt))
summary(g)

data: logpm25, bmicat, NocturnalSympt, logno2_new [554x4]
mapping:  x = logpm25, y = NocturnalSympt
faceting: facet_null() 

No Plot Yet!

g <- ggplot(maacs, aes(logpm25, NocturnalSympt))
print(g)

Error: No layers in plot

这样不会有用

但这样会有

g <- ggplot(maacs, aes(logpm25, NocturnalSympt))
g + geom_point()

Annotation

• Labels: xlab(), ylab(), labs(), ggtitle()
• Each of the “geom” functions has options to modify
• For things that only make sense globally, use theme()
  • Example: theme(legend.position = "none")
• Two standard appearance themes are included
  • theme_gray(): The default theme (gray background)
  • theme_bw(): More stark/plain

alpha就是透明度

此例中因为有y=100，所以才显得第二个图那么变态，第一个图好像少了个头似的，这里只是搞清楚两者的区别即可，但实际应用中有时不需知道全景，只需掌握核心即可

从这图可知道，设置了ylim，会自动去掉outliner,而在后者则不会去掉这个东东。

More Complex Example

• How does the relationship between PM$_{2.5}$ and nocturnal symptoms vary by BMI and NO$_2$?
• Unlike our previous BMI variable, NO2 is continuous
• We need to make NO2 categorical so we can condition on it in the plotting
• Use the cut() function for this

Making NO$_2$ Tertiles

## Calculate the tertiles of the data
cutpoints <- quantile(maacs$logno2_new, seq(0, 1, length = 4), na.rm = TRUE)

## Cut the data at the tertiles and create a new factor variable
maacs$no2tert <- cut(maacs$logno2_new, cutpoints)

## See the levels of the newly created factor variable
levels(maacs$no2tert)

[1] "(0.378,1.2]" "(1.2,1.42]"  "(1.42,2.55]"

Summary

• ggplot2 is very powerful and flexible if you learn the “grammar” and the various elements that can be tuned/modified
• Many more types of plots can be made; explore and mess around with the package (references mentioned in Part 1 are useful)