Presentation of R

An Introduction to R
John Verzani
CUNY/College of Staten Island Department of Mathematics
NYC ASA, CUNY Ed. Psych/May 24, 2005

Outline
What is R?
The many faces of R
Data
Manipulating data
Applying functions to data
Vectorization of data
Graphics
Model formulas
Inference
Significance tests
confidence intervals
Models
Simple linear regression
Multiple linear regression
Analysis of variance models
Logistic regression models
What is R?
R is an open-source statistical computing environment
I R is available from http://www.r-project.org

I R is a computing language, based on S and S-Plus, which is
well suited for statistical calculations
I R has the ability to produce excellent graphics for statistical
explorations and publications
What is R?
The structure of R
Library
Base
base
CLI grid
stats
Batch stats4
...
DCOM,... R Recommended
Kernel library() MASS
nlme
...
Contrib. (CRAN)
UsingR
text graphics
gregmisc
device
...
The many faces of R
The many faces of R
I R is ported to most modern computing platforms: Windows,

MAC OS X, Unix with X11 (linux), ...
I R has an interface that varies depending on the installation:
The many faces of R
Windows interface
Figure: Windows gui

The many faces of R
Mac OS X interface
Figure: Mac OS X gui

The many faces of R
X11 interface – the command line
Figure: There is no standard GUI for X11 implementations. A “typical”

usage may look like this screenshot. Also of interest is ESS package for
(X)Emacs users.
The many faces of R
The command line interface (CLI)
In R the typical means of interacting with the software is at the

command line.
I Commands are typed at the prompt >
I Continuation lines are indicated with a +
I ENTER sends the commands off to the interpreter
2+2
> 2 + 2
[1] 4
Data
Data types
I In statistics data comes in different types: numeric,

categorical, univariate, bivariate, multivariate, etc.
I R has different data types or classes to accommodate these
different types of data sets.
Some basic storage types are:
Data
numeric vectors: created with c(), etc.

> somePrimes = c(2, 3, 5, 7, 11, 13, 17)
> somePrimes
[1] 2 3 5 7 11 13 17
> odds = seq(1, 15, 2)

> odds
[1] 1 3 5 7 9 11 13 15
> ones = rep(1, 10)

> ones
[1] 1 1 1 1 1 1 1 1 1 1
Data
Character strings are indicated by using matching quote, double of

single.
Character variables
> character = c("Homer", "Marge", "Bart", "Lisa",
+ "Maggie")
> gender = c("Male", "Female", "Male", "Female",
+ "Male")
Data
Categorical variables – factors
> gender = factor(c("Male", "Female", "Male",

+ "Female", "Male"))
> gender
[1] Male Female Male Female Male
Levels: Female Male
Factors have an extra attribute a fixed set of levels that require

some care. (E.g., you can’t add new levels without some work.)
Factors are used instead of character vectors, as this allows R to
identify certain types of data. (Storage space is smaller as well.)
Data
Logical vectors: vectors of TRUE or FALSE
> somePrimes
[1] 2 3 5 7 11 13 17
> somePrimes < 10

[1] TRUE TRUE TRUE TRUE FALSE FALSE FALSE
> somePrimes %in% c(3, 5, 7)

[1] FALSE TRUE TRUE TRUE FALSE FALSE FALSE
> somePrimes == 2 | somePrimes >= 10

[1] TRUE FALSE FALSE FALSE TRUE TRUE TRUE
Data
Matrices
defining matrices:matrix(), rbind(), ...
> M = rbind(c(1, 1), c(0, 1))

> M
[,1] [,2]
[1,] 1 1
[2,] 0 1
Data
Matrices, cont.
Operations: multiplication. (* is entry-by-entry
> M %*% M
[,1] [,2]
[1,] 1 2
[2,] 0 1
Inverse is found by “solving” Ax = b, b an indentity matrix
> solve(M)
[,1] [,2]
[1,] 1 -1
[2,] 0 1
Data
Matrices, cont.
Least squares regression coefficients the hard way
> x = 1:5
> y = c(2, 3, 1, 4, 5)
> ones = rep(1, length(x))
> X = cbind(ones, x)
> solve(t(X) %*% X, t(X) %*% y)
[,1]
ones 0.9
x 0.7
Data
Lists
Lists are recursive structures with each level made up of

components.
I List components can be other data types, functions, additional
lists, etc.
I Lists are used often as return values of functions in R. The
print method is set to show only part of the values contained
in the list.
Data
Defining lists
> lst = list(a = somePrimes, b = M, c = mean)

> lst
$a
[1] 2 3 5 7 11 13 17
$b
[,1] [,2]
[1,] 1 1
[2,] 0 1
$c
function (x, ...)
UseMethod("mean")
<environment: namespace:base>
Data
Other data types

Data can be given extra attributes, such as a time series:
Time series have regular date information
> google = c(100.2, 132.6, 196, 180, 202.7,

+ 191.9, 186.1, 180)
> ts(google, start = c(2004, 9), frequency = 12)
Jan Feb Mar Apr May Jun Jul Aug Sep
2004 100.2
2005 202.7 191.9 186.1 180.0
Oct Nov Dec
2004 132.6 196.0 180.0
2005
Data
Tables – an extension of an matrix or array

The table() function (also xtabs, ftable,...)
> table(gender)
gender
Female Male
2 3
> satisfaction = c(3, 4, 3, 5, 4, 3)

> category = c("a", "a", "b", "b", "a", "a")
> table(category, satisfaction)
satisfaction
category 3 4 5
a 2 2 0
b 1 0 1
Data
Data frames
The most common data-storage format is a data frame

I Stores rectangular data: each column a variable, typically each
row data for one subject
I Columns have names for easy reference
I May be manipulated like a matrix or a list (each variable a
top-level component)
Data
Relationship between vector, matrix, data frame, list
Matrix Data frame List
● ● ●
● ● ●
● ● ●
● ● ●
Vector
● ● ●
Data
Data frame examples
> role = c("Comic relief", "Parent", "troublemaker",

+ "Goody two-shoes", "Cute baby")
> theSimpsons = data.frame(name = character,
+ gender = gender, role = role)
> theSimpsons
name gender role
1 Homer Male Comic relief
2 Marge Female Parent
3 Bart Male troublemaker
4 Lisa Female Goody two-shoes
5 Maggie Male Cute baby
Data
Reading in data
Data can be built-in, entered in at the keyboard, or read in from
external files. These may be formatted using fixed width format,
commas separated values, tables, etc. For instance, this command
reads in a data set from a url:
Reading urls
> f = "http://www.math.csi.cuny.edu/st/R/crackers.csv"
> crackers = read.csv(f)
> names(crackers)
[1] "Company" "Product"
[3] "Crackers" "Grams"
[5] "Calories" "Fat.Calories"
[7] "Fat.Grams" "Saturated.Fat.Grams"
[9] "Sodium" "Carbohydrates"
[11] "Fiber"
Data
Manipulating data
Assignment, Extraction
Values in vectors, matrices, lists, and data frames can be accessed
by their components:
By index
> google[1:3]
[1] 100.2 132.6 196.0
> crackers[1:3, 2:3]

Product Crackers
1 Country Water Cracker Crck Pepper 4
2 Country Water Cracker Klassic 4
3 Country Water Cracker Sun Dried Tomato 4
Data
Manipulating data
By index cont.
> M[1, ]
[1] 1 1
> lst[[1]]
[1] 2 3 5 7 11 13 17
Data
Manipulating data
Access by name
by name
> crackers[1:3, c("Product", "Crackers")]

Product Crackers
1 Country Water Cracker Crck Pepper 4
2 Country Water Cracker Klassic 4
3 Country Water Cracker Sun Dried Tomato 4
> theSimpsons[["role"]]
[1] Comic relief Parent troublemaker
[4] Goody two-shoes Cute baby
5 Levels: Comic relief Cute baby ... troublemaker
Data
Manipulating data
Access by logical expressions
logical questions answered TRUE or FALSE
> somePrimes < 10

[1] TRUE TRUE TRUE TRUE FALSE FALSE FALSE
> somePrimes[somePrimes < 10]

[1] 2 3 5 7
Data
Manipulating data
Recycling values
When making assignments in R we might have a situation where
many values are replaced by 1, or a few. R has a means of
recycling the values in the assignment to fill in the size mismatch.
replace coded values with NA
> x = c(1, 1, 0, 1, 99, 0, 1, 99, 0, 1, 99)

> x[x == 99] = NA
> x[x == 1] = "Yes"
> x[x == 0] = "No"
> x
[1] "Yes" "Yes" "No" "Yes" NA "No" "Yes" NA
[9] "No" "Yes" NA
Data
Finding the mean
> fat = crackers$Fat.Grams

> mean(fat)
[1] 3.679
The median
> median(fat)
[1] 3.25
Data
Extra arguments to find trimmed mean
> mean(fat, trim = 0.2)

[1] 3.482
missing data – coded NA
> shuttleFailures = c(0, 1, 0, NA, 0, 0, 0)

> mean(shuttleFailures, na.rm = TRUE)
[1] 0.1667
Data
Functions
I Functions are called by name with a matching pair of ()

I Arguments may be indicated by position or name
I Named arguments can (and usually do) have reasonable
defaults
I A special role is played by the first argument
Data
generic functions
Interacting with R from the command line requires one to
remember a lot of function names, although R helps out
somewhat. In practice, many tasks may be viewed generically:
E.g., “print” the values of an object, “summarize” values of an
object, “plot” the object. Of course, different objects should yield
different representations.
R has methods (S3, S4) to declare a function to be generic. This
allows different functions to be “dispatched” based on the “class” of
the first argument.
A basic template is:
methodName( object, extraArguments)
Some common generic functions are print() (the default action),

summary() (for summaries), plot() (for basic plots).
Data
summary() function called on a number and factor
> summary(somePrimes)
Min. 1st Qu. Median Mean 3rd Qu. Max.
2.00 4.00 7.00 8.29 12.00 17.00
> summary(gender)
Female Male
2 3
Data
R, like MATLAB, is naturally vectorized. For instance, to find the

sample variance, (n − 1)−1 ∑(xi − x̄)2 by hand involves:
sample variance (also var())
> x = c(2, 3, 5, 7, 11, 13)

> fractions(x - mean(x))
[1] -29/6 -23/6 -11/6 1/6 25/6 37/6
> fractions((x - mean(x))^2)

[1] 841/36 529/36 121/36 1/36 625/36 1369/36
> fractions(sum((x - mean(x))^2)/(length(x) -

+ 1))
[1] 581/30
Data
Example: simulating a sample distribution

A simulation of the sampling distribution of x̄ from a random
sample of size 10 taken from an exponential distribution with
parameter 1 naturally lends itself to a “for loop:”
for loop simulation
> res = c()

> for (i in 1:200) {
+ res[i] = mean(rexp(10, rate = 1))
+ }
> summary(res)
0.392 0.786 0.946 1.000 1.220 2.440
Data
Vectorizing a simulation
It is often faster in R to vectorize the simulation above by
generating all of the random data at once, and then applying the
mean() function to the data. The natural way to store the data is
a matrix.
Simulation using a matrix
> m = matrix(rexp(200 * 10, rate = 1), ncol = 200)

> res = apply(m, 2, mean)
> summary(res)
0.308 0.800 1.010 1.020 1.190 1.830
Graphics
Graphics
R has many built-in graphic-producing functions, and facilities to

create custom graphics. Some standard ones include:
histogram and density estimate
> hist(fat, probability = TRUE, col = "goldenrod")

> lines(density(fat), lwd = 3)
Graphics
histogram and density estimate

Histogram of fat
0.20
0.15
Density
0.10
0.05
0.00
0 2 4 6 8
fat
Graphics
Graphics: cont.
Quantile-Quantile plots
> qqnorm(fat)
Boxplots
> boxplot(MPG.highway ~ Type, data = Cars93)
plot() is generic, last one also with
> plot(MPG.highway ~ Type, data = Cars93)

Graphics
Quantile-Quantile plot
Normal Q−Q Plot
●● ● ● ●
8
●●●
●●●●●●●●●●●●
6
Sample Quantiles
●
●●●
●●
●
●●
●
●
●●
●●
●●
●
4
●
●●
●●
●●
●
●●
●●
●●
●●
●●
●●
●●
●
●
●●
●●●
●●●
●
2
●●●●●●●●●●●
● ●●●●●
● ● ●
0
−2 −1 0 1 2
Theoretical Quantiles
Graphics
Boxplots
50 ●
●
45
40
35
30
25
20
Compact Large Midsize Small Sporty Van

Graphics
Fancy examples from upcoming book of P. Murrell

N = 360 brokenness = 0.5
2.0
350
300
1.8
250
234 (65%)
Number of Vessels
Completeness
1.6
200
150
159 (44%)
1.4
100
1.2
1.2
50
1.0
0
0.0 0.2 0.4 0.6 0.8 1.0
Sampling Fraction
Graphics
3-d graphics
X1
X2 X3
X1
X2 X3
Graphics
Model formulas
Model formula notation

The boxplot example illustrates R’s model formula. Many generic
functions have a method to handle this type of input, allowing for
easier usage with multivariate data objects.
Example with xtabs – tables
> df = data.frame(cat = category, sat = satisfaction)

> xtabs(~cat + sat, df)
sat
cat 3 4 5
a 2 2 0
b 1 0 1
Graphics
Model formulas
Model formula cont.
Suppose x, y are numeric variables, and f is a factor. The basic

model formulas have these interpretations:
y ~ 1 yi = β0 + εi
y ~ x yi = β0 + β1 xi + εi
y ~ x - 1 yi = β1 xi + εi remove the intercept
y ~ x | f yi = β0 + β1 xi + εi grouped by levels of f
y ~ f yij = τi + εij
The last usage suggests storing multivariate data in two variables –

a numeric variable with the measurements, and a factor indicating
the treatment.
Graphics
Model formulas
Lattice graphics
Lattice graphics can effectively display multivariate data that are

naturally defined by some grouping variable. (Slightly more
complicated than need be to show groupedData() function in
nlme package.)
lattice graphics
> cars = groupedData(MPG.highway ~ Weight |

+ Type, Cars93)
> plot(cars)
Graphics
Model formulas
2000 3000 4000
Compact Sporty Small

● 50
●
45
●
●
40
● ● ● ●●●
●● ● 35
● ●●
●● ●
● ●●● ●● ●
●●●
● ●● ● ●●●● 30
●● ●● ●
MPG.highway
● ● ●● ● ● 25
20
Van Large Midsize
50
45
40
35
●
30 ● ●●● ●
●
●
● ● ●● ●●●●
●● ●
●●● ●
25 ● ● ●
●● ●
●● ● ● ●
●
● ●●
20
2000 3000 4000 2000 3000 4000

Weight
Graphics
Model formulas
more lattice graphics (Murrell’s book)
Waseca
Trebi ●
Wisconsin No. 38 ●
No. 457 ●
Glabron ●
Peatland ●
Velvet ●
No. 475 ●
Manchuria ●
No. 462 ●
Svansota ●
Crookston
Trebi ●
No. 457 ●
Glabron ●
Peatland ●
Velvet ●
No. 475 ●
Manchuria ●
No. 462 ●
Svansota ●
Morris
Trebi ●
No. 457 ●
Glabron ●
Peatland ●
Velvet ●
No. 475 ●
Manchuria ●
No. 462 ●
Svansota ● 1932 ●
University Farm 1931
Trebi ●
No. 457 ●
Glabron ●
Peatland ●
Velvet ●
No. 475 ●
Manchuria ●
No. 462 ●
Svansota ●
Duluth
Trebi ●
No. 457 ●
Glabron ●
Peatland ●
Velvet ●
No. 475 ●
Manchuria ●
No. 462 ●
Svansota ●
Grand Rapids
Trebi ●
No. 457 ●
Glabron ●
Peatland ●
Velvet ●
No. 475 ●
Manchuria ●
No. 462 ●
Svansota ●
20 30 40 50 60
Barley Yield (bushels/acre)

Inference
Significance tests
Significance tests
There are several functions for performing classical statistical tests

of significance: t.test(), prop.test(), oneway.test(),
wilcox.test(), chisq.test(), ...
These produce a p-value, and summaries of the computations.
Inference
Significance tests
The Bumpus data set (Ramsey and Shafer) contains data from
1898 lecture supporting evolution (Some birds survived a harsh
winter storm)
two-sample t test
> Bumpus = read.table("Bumpus.txt", header = TRUE)

> plot(humerus ~ factor(code), data = Bumpus)
Inference
Significance tests
Diagnostic plot
780
760
740
humerus
720
700
●
680
660
1 2
factor(code)
Inference
Significance tests
t.test() output
> t.test(humerus ~ code, data = Bumpus)
Welch Two Sample t-test
data: humerus by code

t = -1.721, df = 43.82, p-value = 0.09236
alternative hypothesis: true difference in means is not equ
95 percent confidence interval:
-21.895 1.728
sample estimates:
mean in group 1 mean in group 2
727.9 738.0
Inference
Significance tests
The SchizoTwins data set (R&S) contains data on 15 pairs of

monozygotic twins. Measured values are of volume of left
hippocampus.
t-tests: paired
> twins = read.table("SchizoTwins.txt", header = TRUE)

> plot(affected ~ unaffected, data = twins)
> attach(twins)
> t.test(affected - unaffected)$p.value
[1] 0.006062
> t.test(affected, unaffected, paired = TRUE)$p.value
[1] 0.006062
> detach(twins)
Inference
Significance tests
Diagnostic plot
2.0 ●
●
●
●
1.8
●
●
●
1.6
●
affected
●
1.4
●
●
● ●
●
1.2
●
1.0
1.4 1.6 1.8 2.0
unaffected
Inference
Confidence intervals
Confidence intervals are computed as part of the output of many

of these functions. The default is to do 95% CIs, which may be
adjusted using conf.level=.
95% CI humerus length overall
> t.test(Bumpus$humerus)
...
95 percent confidence interval:
728.2 739.6
...
Inference
Chi-square tests
Goodness of fit tests are available through chisq.test() and
others. For instance, data from Rosen and Jerdee (1974, from
R&S) on the promotion of candidates based on gender:
gender data
> rj = rbind(c(21, 3), c(14, 10))

> dimnames(rj) = list(gender = c("M", "F"),
+ promoted = c("Y", "N"))
> rj
promoted
gender Y N
M 21 3
F 14 10
Inference
sieveplot(rj)
Sieve diagram
F
gender
M
Y N
promoted
Inference
chi-squared test p-value
> chisq.test(rj)$p.value
[1] 0.05132
Fischer’s exact test

> fisher.test(rj, alt = "greater")$p.value
[1] 0.02450
Models
Fitting linear models
Linear models are fit using lm():

I This function uses the syntax for model formula
I Model objects are reticent – you need to ask them for more
information
I This is done with extractor functions: summary(), resid(),
fitted(), coef(), predict(), anova(), deviance(),...
Models
Body fat data set
> source("http://www.math.csi.cuny.edu/st/R/fat.R")
> names(fat)
[1] "case" "body.fat" "body.fat.siri"
[4] "density" "age" "weight"
[7] "height" "BMI" "ffweight"
[10] "neck" "chest" "abdomen"
[13] "hip" "thigh" "knee"
[16] "ankle" "bicep" "forearm"
[19] "wrist"
Models
Fitting a simple linear regression model

Basic fit is done with lm: response ˜ predictor(s)
> res = lm(body.fat ~ BMI, data = fat)

> res
Call:
lm(formula = body.fat ~ BMI, data = fat)
Coefficients:
(Intercept) BMI
-20.41 1.55
Models
Making scatterplot, adding the regression line
> plot(body.fat ~ BMI, data = fat)

> abline(res)
> res.robust = lqs(body.fat ~ BMI, data = fat)
> abline(res.robust, lty = 2, col = "blue")
> title("BMI predicting body fat")
> legend(35, 20, legend = c("least-squares",
+ "lqs"), lty = 1:2)
Models
Scatterplot with regression line

BMI predicting body fat
40 ●
●
●
●
● ● ●
●
●●
● ● ● ●●
● ●●● ● ●
30
● ● ●● ● ●
● ● ●● ● ● ●●
● ●
● ●●●
●● ●●● ●●● ● ● ●
body.fat
●●● ● ● ● ●
●●●●●● ● ●● ● ●●
● ● ● ●● ●
●● ●● ● ● ●
●●●●
● ● ●●
● ● ●
●● ● ● ● ● ●
● ● ●
20
● ●● ●●●●● ● ● ●
● ●● ●●
● ● ● ●● ●
● ● ●● ●● ● ●
●
●●●● ●●● ●
● least−squares
● ●●●●●
● ●● ●
● ●● ●
●
● ●
●● ●● ●
●● ●
●● ● lqs
● ●
●●
●
● ●● ● ●
●● ●●●
● ● ● ●● ●●● ●
●
● ●●●● ●● ●
10
● ●● ●●●● ● ●
●● ●●●
●
● ● ●
●● ●
●● ● ●● ● ●
● ● ●
● ● ●● ●
●
●
●
0
20 25 30 35 40 45 50
BMI
Models
Basic output is minimal, more is given by summary()
> summary(res)
Call:
lm(formula = body.fat ~ BMI, data = fat)
Residuals:
Min 1Q Median 3Q Max
-21.4292 -3.4478 0.2113 3.8663 11.7826
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -20.40508 2.36723 -8.62 7.78e-16 ***
BMI 1.54671 0.09212 16.79 < 2e-16 ***
---
Signif. codes: 0 ’***’ 0.001 ’**’ 0.01 ’*’ 0.05 ’.’ 0.1 ’
Models
summary() output cont.
Residual standard error: 5.324 on 250 degrees of freedom

Multiple R-Squared: 0.53, Adjusted R-squared: 0.5281
F-statistic: 281.9 on 1 and 250 DF, p-value: < 2.2e-16
Models
Extractor functions used to extract information

extractor functions
> coef(res)
(Intercept) BMI
-20.405 1.547
> summary(residuals(res))
Min. 1st Qu. Median Mean 3rd Qu.
-2.14e+01 -3.45e+00 2.11e-01 3.52e-17 3.87e+00
Max.
1.18e+01
Models
Residual plots to test model assumptions
> plot(fitted(res), resid(res))

> qqnorm(resid(res))
Models
Residual plots
Normal Q−Q Plot
● ● ● ● ● ●
● ●
10
10
● ● ●●
● ● ● ●●●
● ●● ● ●●●●
● ● ● ●
●●
● ●●● ● ● ● ● ●
●●
●●
●●
●
●● ● ● ● ●
●●
●●
●
●● ● ● ●● ● ●● ●
●●
●
●●
●
●●
●● ● ● ● ● ● ●● ●
●
●●
●
●●
●
● ●● ●
●
●●
5
5
● ● ● ● ●
●
●
●●
●● ●●●● ● ●
●
●
● ● ●● ● ●● ●
● ● ● ●
●●
●
●
●
●
●●● ●● ● ● ●●
●
●
●
●
●
●● ● ● ● ● ● ●● ●●● ●
●●
●
●●
● ● ● ●
●● ●● ● ● ● ● ● ●
●●
●
●●
●● ●
● ● ● ● ● ●●
●
●●
● ● ● ● ●●●● ● ●●
Sample Quantiles
●● ● ●
●●
●
0
0
● ●
● ● ●● ● ●● ●
● ● ●
● ●●
●
●
●
●
●
● ●
● ●● ● ● ●● ●●
● ● ●● ●● ●● ● ● ●
●
●
●
●
●●● ● ● ● ● ●
●
resid(res)
● ● ● ●
●●
●●
●● ● ● ● ● ● ● ●●
●●
● ● ●●●● ●● ●● ● ●● ● ●
●●
●●
●
● ●● ● ● ● ● ●
●
●●
●
●●
−5
−5
● ● ● ● ●● ●●
●
●
●●
● ● ● ● ●● ● ●
●●
●
●
●●
●
● ●● ● ●
●
●●
● ●● ● ●● ● ● ●
●●
●
●
●
●
●
● ●● ● ●
●
●
● ● ● ● ● ●
●●
●●
●
● ● ● ● ●
●●●●
●●
−10
−10
● ● ● ● ●●●●●
● ● ●●
● ●
−15
−15
−20
●
−20 ●
10 20 30 40 50 −3 −2 −1 0 1 2 3
fitted(res) Theoretical Quantiles

Models
Default diagnostic plots
Model objects, such as the output of lm(), have default plots

associated with them. For lm() there are four plots.
plot(res)
> par(mfrow = c(2, 2))

> plot(res)
Models
Diagnostic plots for lm()
Residuals vs Fitted Normal Q−Q plot
Standardized residuals
●●●
81 81●
●●
10 ● ● ● ●●●
2
●● ● ●
● ●●● ●●
●
●
●
●
●●●● ●●● ● ●
●●
●
●
●●
●
●●
●
●●
●●
● ●
● ●●
●
●
●●● ●●●
● ●
●●
●
●
●●●●●●●●
●
● ● ●
●●
●
●
●●
●
●●
●
●●●●●
●●●● ●
●●●●●● ●● ● ●
●
●●
●●
●
●● ●
Residuals
●●● ●●
● ●● ●●● ●●●●
●●●●
●● ●
●
●
●●
●
●
●●● ●●●●● ●●
●
●●●●●● ● ●● ●●
●
●
●●
●
●
●●● ●● ● ●
●
0
0
●● ● ●
●
● ●●●
●●● ● ●
●●
●
●
●●
●●● ● ●●●● ● ●● ● ●
●
●
●●
●● ●●●●
●●
●
●
●
●
●
●
●●
●
● ● ●●
●●
● ●
●●
●●●
●●● ●● ●
●
●
●
●
●
●●
●
●●
●
●
●● ● ● ● ● ● ● ●
●
●●
●
●
●●●
●●●●
● ● ● ● ●
●
●●
● ●● ● ●●
●
●
−20 −10
● ●
● ● ● ●● ●
●
●●
●
●
●●
● ●● ● ● ●●
●●
●
●
−2
●●
●● ● ●●
9 9●
−4
39●
39●
10 20 30 40 50 −3 −2 −1 0 1 2 3
Fitted values Theoretical Quantiles
Scale−Location plot Cook's distance plot

Standardized residuals
39●
0.0 0.5 1.0 1.5 2.0
0.0 0.5 1.0 1.5 2.0

39
Cook's distance
9●●●●
●
81 ●
●● ●●●● ●● ●● ●
●● ●●
●●●
● ●●● ● ●
● ●● ●
● ●● ●●● ● ●●
●●●●
● ●
●
● ●●
●●
●●● ●
●● ●● ●●●●●● ●● ●●
●
●● ●● ●
●
●● ●● ●●●● ●● ●
●
●●●●●●●●
●
●
●●●
● ●● ● ●●
●● ●●● ●● ●
●●● ●●
●●
●●● ●●●
●
●
●
●●
●● ●●
●● ●● ●●
●● ●
●●
● ●● ●
●●
●●●
● ●●● ●
● ●●●● ●
●
●●
●●
● ● ●●
● ●● ●● ●●●● ●
● ●●● ●● ●
●
●● ● ● ●●●● ●● ●
●●● ●● ●●●● ●
● 41 216
10 20 30 40 50 0 50 100 150 200 250
Fitted values Obs. number

Models
Predictions done using the predict() extractor function
> vals = seq(15, 35, by = 2)

> names(vals) = vals
> predict(res, newdata = data.frame(BMI = vals))
15 17 19 21 23 25 27
2.796 5.889 8.982 12.076 15.169 18.263 21.356
29 31 33 35
24.450 27.543 30.636 33.730
Models
Multiple regression
Multiple regression models are modeled with lm() as well. Extra

covariates are specified using the following notations:
I + adds terms (- subtracts them, such as -1)
I Math expressions can (mostly) be used as is: log, exp,...
I I() used to insulate certain math expressions
I a:b adds an interaction between a and b. Also, *, ^ are
shortcuts for more complicated interactions
To illustrate, we model the body.fat variable, by measurements
that are easy to compute
Models
Modeling body fat
> res = lm(body.fat ~ age + weight + height +

+ chest + abdomen + hip + thigh, data = fat)
> res
Call:
lm(formula = body.fat ~ age + weight + height + chest + abd
Coefficients:
(Intercept) age weight height
-33.27351 0.00986 -0.12846 -0.09557
chest abdomen hip thigh
-0.00150 0.89851 -0.17687 0.27132
Models
Model selection using AIC
> stepAIC(res, trace = 0)
Call:
lm(formula = body.fat ~ weight + abdomen + thigh, data = fa
Coefficients:
(Intercept) weight abdomen thigh
-48.039 -0.170 0.917 0.209
(Set trace=1 to get diagnostic output.)

Models
Model selection using F -statistic
> res.sub = lm(body.fat ~ weight + height +

+ abdomen, fat)
> anova(res.sub, res)
Analysis of Variance Table
Model 1: body.fat ~ weight + height + abdomen

Model 2: body.fat ~ age + weight + height + chest + abdomen
Res.Df RSS Df Sum of Sq F Pr(>F)
1 248 4206
2 244 4119 4 88 1.3 0.27
Models
A simple one-way analysis of variance test is done using,

oneway.test() with a model formula of the type y ~ f.
For instance, we look at data on the lifetime of mice who have
been given a type of diet (R&S).
read in data, make plot
> mice = read.table("mice-life.txt", header = TRUE)

> plot(lifetime ~ treatment, data = mice, ylab = "Months")
Models
Plot of months survived by diet
50
40
Months
30
●
●
●
20
●
● ●
●
●
●
10
N/N85 N/R40 N/R50 NP R/R50 lopro
treatment
Models
One way test of equivalence of means
> oneway.test(lifetime ~ treatment, data = mice,

+ var.equal = TRUE)
One-way analysis of means
data: lifetime and treatment

F = 57.1, num df = 5, denom df = 343, p-value <
2.2e-16
(var.equal=TRUE for assumption of equal variances, not default)

Models
Using lm() for ANOVA

Modeling is usually done with a modeling function. The lm()
function can also fit a one-way ANOVA, again with the same
model formula
Using lm()
> res = lm(lifetime ~ treatment, data = mice)

> anova(res)
Analysis of Variance Table
Response: lifetime
Df Sum Sq Mean Sq F value Pr(>F)
treatment 5 12734 2547 57.1 <2e-16 ***
Residuals 343 15297 45
---
Signif. codes: 0 ’***’ 0.001 ’**’ 0.01 ’*’ 0.05 ’.’ 0.1 ’
Models
Following Ramsey and Schafer we ask: Does lifetime on 50kcal/wk

exceed that of 85 kcal/month? We can take advantage of the use
of treatment contrasts to investigate this (difference in mean from
first level is estimated).
Treatment contrasts, set β1
> treat = relevel(mice$treatment, "N/R50")

> res = lm(lifetime ~ treat, data = mice)
> coef(summary(res))
Estimate Std. Error t value Pr(>|t|)
(Intercept) 42.30 0.8 53.37 3.4e-168
treatN/N85 -9.61 1.2 -8.09 1.1e-14
treatN/R40 2.82 1.2 2.41 1.7e-02
treatNP -14.90 1.2 -12.01 5.7e-28
treatR/R50 0.59 1.2 0.49 6.2e-01
treatlopro -2.61 1.2 -2.19 2.9e-02
Models
logistic regression
Logistic regression extends the linear regression framework to

binary response variables. It may be seen as a special case of a
generalized linear model which consists of:
I A response y and covariates x1 , x2 , . . . , xp
I A linear predictor η = β1 x1 + · · · + βp xp .
I A specific family of distributions which describe the random
variable y with mean response, µy |x , related to η through a
link function, m−1 , where
µy |x = m(η), or η = m−1 (µy |x )

Models
logistic regression example
I Linear regression would be m being the identity and the

distribution being the normal distribution.
I For logistic regression, the response variable is Bernoulli, so
the mean is also the probability of success. The link function
is the logit, or log-odds, function and the family is Bernoulli, a
special case of the binomial.
We illustrate the implementation in R with an example from
Devore on the failure of space shuttle rings (a binary response
variable) in terms of take off temperature.
Models
Space shuttle data
read in data, diagnostic plot
> shuttle = read.table("shuttle.txt", header = TRUE)

> dotplot(~Temperature | Failure, data = shuttle,
+ layout = c(1, 2))
Models
Lift-off temperature by O-Ring failure/success (Y/N)
● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ●
55 60 65 70 75 80
Temperature
Models
Fitting a logistic model

We need to specify the formula, the family, and the link to the
glm() function:
Specify model, family, optional link
> res.glm = glm(Failure ~ Temperature, data = shuttle,

+ family = binomial(link = logit))
> coef(summary(res.glm))
Estimate Std. Error z value Pr(>|z|)
(Intercept) 11.7464 6.0214 1.951 0.05108
Temperature -0.1884 0.0891 -2.115 0.03443
(Actually, link=logit is default for binomial family.)

Models
Random effects models
mixed effects
Mixed-effects models are fit using the nlme package (or its new
replacement lmer which can also do logistic regression).
I Need to specify the fixed and random effects
I Can optionally specify structure beyond independence for the
error terms.
I The implemention is well documented in Pinheiro and Bates
(2000)
Models
Mixed-effects example
We present an example from Piheiro and Bates on a data set

involving a dental measurement taken over time for the same set of
subjects – longitudinal data.
The key variables
I Orthodont the data frame containing:
I distance – measurement
I age – age of subject at time of measurement
I Sex – gender of subject
I subject – subject code
Models
Fit model with lm(), check

The simple regression model is
yi = β0 + β1 xi + εi ,
where εi is N (0, σ2 ).
model using lm()
> res.lm = lm(distance ~ I(age - 11), Orthodont)

> res.lm
> bwplot(getGroups(Orthodont) ~ resid(res.lm))
Call:
lm(formula = distance ~ I(age - 11), data = Orthodont)
Coefficients:
(Intercept) I(age - 11)
24.02 0.66
Models
Boxplots of residuals by group
F11 ●
F04 ●
F03 ●
F08 ●
F02 ●
F07 ●
F05 ●
F01 ●
F06 ●
F09 ●
F10 ●
M10 ●
M01 ●
M04 ●
M06 ●
M15 ●
M09 ●
M14 ●
M13 ●
M12 ●
M03 ●
M08 ●
M07 ●
M11 ●
M02 ●
M05 ●
M16 ●
−5 0 5
resid(res.lm)
Models
Fit each group
A linear model fit for each group is this model
yij = βi0 + βi1 xij + εij ,
where εij are N (0, σ2 ).
fit each group with lmList()
> res.lmlist = lmList(distance ~ I(age - 11) |

+ Subject, data = Orthodont)
> plot(intervals(res.lmlist))
Models
Intervals of slope, intercept estimate by group

(Intercept) I(age − 11)
F11 | | | | | |
F04 | | | | | |
F03 | | | | | |
F08 | | | | | |
F02 | | | | | |
F07 | | | | | |
F05 | | | | | |
F01 | | | | | |
F06 | | | | | |
F09 | | | | | |
F10 | | | | | |
M10 | | | | | |
M01 | | | | | |
Subject
M04 | | | | | |
M06 | | | | | |
M15 | | | | | |
M09 | | | | | |
M14 | | | | | |
M13 | | | | | |
M12 | | | | | |
M03 | | | | | |
M08 | | | | | |
M07 | | | | | |
M11 | | | | | |
M02 | | | | | |
M05 | | | | | |
M16 | | | | | |
20 25 30 −0.5 0.0 0.5 1.0 1.5 2.0 2.5

Models
Fit with random effect
This fits the model
yij = (β0 + b0i ) + (β1 + b1i )xij + εij ,
with b·i N (0, Ψ). εij N (0, σ2 ).
fit with lme()
> res.lme = lme(distance ~ I(age - 11), data = Orthodont,

+ random = ~I(age - 11) | Subject)
> plot(augPred(res.lme))
> plot(res.lme, resid(.) ~ fitted(.) | Sex)
Models
Predictions based on random-effects model

8 9 10 11 12 13 14
F03 F04 F11

30
● ●
● ●
25 ● ● ● ● ●
● ●
20 ●
F06 F01 F05 F07 F02 F08

30
● ● 25
Distance from pituitary to pterygomaxillary fissure (mm)
● ● ● ●
● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ●
● ● 20
M06 M04 M01 M10 F10 F09

● ● ●
30
● ●
● ● ●
● ● ●
● ● ●
25 ● ●
● ●
●
20
● ● ● ●
M03 M12 M13 M14 M09 M15

●
● ● 30
● ●
● ● ● ● ● ●
● ● 25
● ● ●
● ● ● ● ●
●
● 20
●
M16 M05 M02 M11 M07 M08

30
● ● ●
25 ● ● ●
● ● ●
● ● ● ● ● ●
● ● ● ● ●
● ● ●
20 ●
8 9 10 11 12 13 14 8 9 10 11 12 13 14 8 9 10 11 12 13 14
Age (yr)
Models
Residuals by gender, subject

20 25 30
Male Female
2
● ● ●
●
● ●
● ●
●● ● ● ● ● ● ● ●
Residuals (mm)
● ● ● ● ● ●
● ● ●● ● ● ●
● ● ●
● ● ●
●● ● ● ●
● ● ● ●
0 ●
●● ●● ●● ● ●
●● ● ●● ● ●
●
●
●● ●
● ● ● ● ● ● ●
● ●
● ● ●
● ● ● ●●● ●
● ● ●
● ●●● ● ● ●● ●
● ●
● ●
●
−2
●
−4
●
20 25 30
Fitted values (mm)

Models
Adjust the variance
Adjust the variance for different groups is done by specifying a

formula to the weights= argument:
Adjust σ for each gender
> res.lme2 = update(res.lme, weights = varIdent(form = ~1 |

+ Sex))
> plot(compareFits(ranef(res.lme), ranef(res.lme2)))
> plot(comparePred(res.lme, res.lme2))
Models
Compare random effects BLUPs for two models

● ranef(res.lme) ranef(res.lme2)
(Intercept) I(age − 11)

F11 ● ●
F04 ● ●
F03 ● ●
F08 ● ●
F02 ● ●
F07 ● ●
F05 ● ●
F01 ● ●
F06 ● ●
F09 ● ●
F10 ● ●
M10 ● ●
M01 ● ●
M04 ● ●
M06 ● ●
M15 ● ●
M09 ● ●
M14 ● ●
M13 ● ●
M12 ● ●
M03 ● ●
M08 ● ●
M07 ● ●
M11 ● ●
M02 ● ●
M05 ● ●
M16 ● ●
−4 −2 0 2 4 −0.2 0.0 0.2 0.4

Models
Compare the different predictions

res.lme res.lme2
8 9 10 11 12 13 14
F03 F04 F11

30
● ●
● ●
25 ● ● ● ● ●
● ●
20 ●
F06 F01 F05 F07 F02 F08

30
●
Distance from pituitary to pterygomaxillary fissure (mm)
● 25
● ● ● ●
● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ●
● ● 20
M06 M04 M01 M10 F10 F09

● ● ●
30
● ●
● ● ●
● ● ●
● ● ●
25 ● ●
● ●
●
20
● ● ● ●
M03 M12 M13 M14 M09 M15

●
● ● 30
● ●
● ● ● ● ● ●
● ● 25
● ● ●
● ● ● ● ●
●
● 20
●
M16 M05 M02 M11 M07 M08

30
● ● ●
25 ● ● ●
● ● ●
● ● ● ● ● ●
● ● ● ● ●
● ● ●
20 ●
8 9 10 11 12 13 14 8 9 10 11 12 13 14 8 9 10 11 12 13 14
Age (yr)
Models
compare models using anova()

These nested models can be formally compared using a likelihood
ratio test. The details are carried out by the anova() method:
compare nested models
> anova(res.lme, res.lme2)

Model df AIC BIC logLik Test L.Ratio
res.lme 1 6 454.6 470.6 -221.3
res.lme2 2 7 435.6 454.3 -210.8 1 vs 2 20.99
p-value
res.lme
res.lme2 <.0001
Extras
Extending R: writing functions
R can be extended by writing functions. These may be defined in

separate files and read into R, or defined within an R session. For
instance, how to create the following diagram?
Extras
Pizza and pepperoni

Extras
helper functions
plotCircle = function(x,radius=1,...) {
t = seq(0,2*pi,length=100)
polygon(x[1]+ radius*cos(t), x[2]+radius*sin(t), ...)
}
doPlot = function(x,r,R,...) {
if(sqrt(sum(x^2))+r < R) plotCircle(x,r=r,...)
}
Extras
Key points
I functions are defined using function()

I Arguments are matched by position or name
I Named arguments may be abbreviated
I The special argument ... is used to pass along extra
arguments
I Command blocks are indicated using braces
I Functions can be defined on the command line, used
anonymously, or be stored in files to be sourced in.
Extras
Calling function
plotPizza = function(n,R=n,r=0.2) {
par(mai=c(0,0,0,0))
plot.new();plot.window(xlim=c(-n,n),ylim=c(-n,n),asp=1)
plotCircle(c(0,0),radius=R,lwd=2)
x = rep(-n:n,rep(2*n+1,2*n+1))
y = rep(-n:n,length.out=(2*n+1)^2)
apply(cbind(x,y),1,function(x) doPlot(x,r,R,col=gray(.5))
}
plotPizza(5)
Extras
Extending R with add-on packages
Extending R: add-on packages
One can extend R using add-on packages. Some are built-in

(≈ 10), over 400 contributed packages are hosted on CRAN
(http://cran.r-project.org), others are on author’s websites.
For instance, neglecting issues with permissions, to download and
install a basic GUI for R can be done with the command
Installing a package
> install.packages("Rcmdr")
This GUI, available for the three main platforms, allows one to
select variables, and fill in function arguments with a mouse.
The command install.packages() allows one to browse the
available packages.
Extras
Learning more
R has several different ways that you can learn more:

The built in help pages. Some key fuctions are
I help.start() – to start web interface
I ?functionName – to find specific help for the named function
I apropos("word") To search through searchlist for a word
I help.search() matches in more places than apropos()
Extras
More free documentation
The accompanying manuals Included with R are 5 manuals in pdf

or html form. An Introduction to R contains lots of
information.
Contributed documentation Contributed documentation: On the R
project webpage http://www.r-project.org is a
link to contributed documentation. There are quite a
few documents of significant size, including a few
that have made it into book form.
The R mailing list The R mailing list is full of information.
Questions should only be asked after a reading of the
FAQ or you are likely to have a rather terse response.
Extras
Books on R (Also numerous S-plus titles)
Verzani, Using R for Introductory Statistics
Dalgaard, Introductory Statistics with R
Maindonald and Braun, Data Analysis and Graphics Using R
Fox, An R and S−Plus Companion to Applied Regression
Faraway, Linear Models with R
Heiberger and Holland, Statistical Analysis and Data Display...
Venables and Ripley, Modern Applied Statistics with S
Pinheiro and Bates, Mixed−Effects Models in S and S−Plus
Venables and Ripley, S Programming
Chambers and Hastie, Statistical Models in S.
Chambers, Programming with Data
Paul Murrell, R Graphics
200 400 600

Presentation of R

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Presentation of R

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Presentation of R

Uploaded by

Copyright:

Available Formats

An Introduction to R

CUNY/College of Staten Island Department of Mathematics

NYC ASA, CUNY Ed. Psych/May 24, 2005

R is an open-source statistical computing environment

I R is available from http://www.r-project.org

The many faces of R

I R is ported to most modern computing platforms: Windows,

Figure: Windows gui

Figure: Mac OS X gui

X11 interface – the command line

Figure: There is no standard GUI for X11 implementations. A “typical”

The command line interface (CLI)

In R the typical means of interacting with the software is at the

I In statistics data comes in different types: numeric,

numeric vectors: created with c(), etc.

> odds = seq(1, 15, 2)

> ones = rep(1, 10)

Character strings are indicated by using matching quote, double of

Categorical variables – factors

> gender = factor(c("Male", "Female", "Male",

Factors have an extra attribute a fixed set of levels that require

Logical vectors: vectors of TRUE or FALSE

> somePrimes < 10

> somePrimes %in% c(3, 5, 7)

> somePrimes == 2 | somePrimes >= 10

defining matrices:matrix(), rbind(), ...

> M = rbind(c(1, 1), c(0, 1))

Inverse is found by “solving” Ax = b, b an indentity matrix

Least squares regression coefficients the hard way

Lists are recursive structures with each level made up of

> lst = list(a = somePrimes, b = M, c = mean)

Other data types

> google = c(100.2, 132.6, 196, 180, 202.7,

Tables – an extension of an matrix or array

> satisfaction = c(3, 4, 3, 5, 4, 3)

The most common data-storage format is a data frame

Relationship between vector, matrix, data frame, list

Matrix Data frame List

Data frame examples

> role = c("Comic relief", "Parent", "troublemaker",

> crackers[1:3, 2:3]

> crackers[1:3, c("Product", "Crackers")]

Access by logical expressions

logical questions answered TRUE or FALSE

> somePrimes < 10

> somePrimes[somePrimes < 10]

> x = c(1, 1, 0, 1, 99, 0, 1, 99, 0, 1, 99)

Applying functions to data

Finding the mean

> fat = crackers$Fat.Grams

Extra arguments to find trimmed mean

> mean(fat, trim = 0.2)

missing data – coded NA

> shuttleFailures = c(0, 1, 0, NA, 0, 0, 0)

I Functions are called by name with a matching pair of ()

Some common generic functions are print() (the default action),

summary() function called on a number and factor

R, like MATLAB, is naturally vectorized. For instance, to find the

> x = c(2, 3, 5, 7, 11, 13)