Package prob

February 16, 2017

Version 1.0-0
Date 2017-02-15
Title Elementary Probability on Finite Sample Spaces
Depends combinat, fAsianOptions
Suggests VGAM, reshape, MASS, hypergeo
Description A framework for performing elementary probability
calculations on finite sample spaces, which may be represented by data frames
or lists. Functionality includes setting up sample spaces, counting tools,
defining probability spaces, performing set algebra, calculating probability
and conditional probability, tools for simulation and checking the law of
large numbers, adding random variables, and finding marginal distributions.
Characteristic functions for all base R distributions are included.
License GPL (>= 3)

URL http://prob.r-forge.r-project.org, http://gkerns.people.ysu.edu/

NeedsCompilation no
Author G. Jay Kerns [aut, cre, cph]
Maintainer G. Jay Kerns <[email protected]>
Repository CRAN
Date/Publication 2017-02-16 07:13:39

R topics documented:
prob-package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
addrv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
cards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
CharFunc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
countrep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
empirical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
euchredeck . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
gen2wayTable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
genIndepTable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

1
2 prob-package

genLogRegData . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
genXdata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
iidspace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
intersect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
is.probspace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
isin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
isrep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
marginal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
noorder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
nsamp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
permsn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
prob . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
probspace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
rolldie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
roulette . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
setdiff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
sim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
subset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
tosscoin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
union . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
urnsamples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Index 36

prob-package Elementary Probability on Finite Sample Spaces

Description
A framework for performing elementary probability calculations on finite sample spaces. It is built
around the concept of a probability space, which is an object of outcomes and an object probs of
probabilities associated with the outcomes.
There are two ways to represent a probability space in the prob package. The first is with a data
frame that has a probs column. Entries of probs should be nonnegative and sum to one. The
second way is with a list having two components: outcomes and probs. The component outcomes
is a list containing elements of the most arbitrary sort; they can be data frames, vectors, more lists,
whatever. The probs component is a vector (of the same length as outcomes), which associates to
each element of outcomes a nonnegative number. As before, the only additional requirement is that
the sum of probs be one.
There are functions in the prob package to address many topics in a standard course in elementary
probability. In particular, there are methods for setting up sample spaces, counting tools, defin-
ing probability spaces, performing set algebra, calculating probability and conditional probability,
tools for simulation and checking the law of large numbers, adding random variables, and finding
marginal distributions. See vignette("prob") for details.
There are some functions included to set up some of the standard sample spaces usually encountered
in an elementary probability course. Examples include tossing a coin, rolling a die, drawing from a
52 card deck, etc. If you know of topics that would be of general interest and could be incorporated
addrv 3

in the prob package framework, I would be happy to hear about them. Comments and suggestions
are always welcomed.
The prob package is a first step toward addressing probability in R, and has been written in the spirit
of simplicity. The procedures work best to solve problems that are manageable in scope. Users that
wish to investigate especially large or intricate problems are encouraged to modify and streamline
the code to suit their individual needs.
Characteristic functions for the base probability distributions have been included. For details, type
vignette("charfunc") at the command prompt.

Details

Package: prob
Version: 1.0-0
Date: 2017-02-15
Depends: combinat, fAsianOptions
Suggests: VGAM, hypergeo
LazyLoad: no
License: GPL version 3 or newer
URL: http://prob.r-forge.r-project.org, http://gkerns.people.ysu.edu/

Author(s)

G. Jay Kerns <[email protected]>

Maintainer: G. Jay Kerns <[email protected]>

addrv Adding Random Variables to a Probability Space

Description

Adds a column to a data frame probability space containing the values of a random variable com-
puted from the existing columns of the space.

Usage

addrv(space, FUN = NULL, invars = NULL, name = NULL, ...)

4 cards

Arguments
space a data frame with a probs column.
FUN a function to be applied to each row of outcomes in space
invars a character vector indicating input columns of space
name an (optional) name to give the defined random variable.
... an expression defining a random variable.

Details
There are two ways to add a random variable to a data frame probability space; see the examples.
The argument FUN has precedence and will be used if specified. If name is not specified, then the new
random variable will be called X. Note that this function only works for data frames, as a method
for objects of class ps has not been implemented.

Value
The input data frame with an additional column called name.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
See transform to add a column to a data frame of outcomes (not yet a probability space).

Examples
S <-rolldie(3, makespace = TRUE)
addrv(S, sum, name = "Y")
addrv(S, Z = X3 - X2 )

cards A Standard Set of Playing Cards

Description
The title says it all.

Usage
cards(jokers = FALSE, makespace = FALSE)
CharFunc 5

Arguments
jokers logical.
makespace logical.

Details
This generates a data frame sample space of a standard deck of 52 playing cards. Optionally, the
user can specify that Jokers be included, which have a rank but with suit a missing value.

Value
A data frame with columns rank and suit, and optionally a column of equally likely probs.

See Also
rolldie, tosscoin, and roulette

Examples
cards()
cards(makespace = TRUE)

CharFunc Characteristic functions

Description
The characteristic functions for selected probability distributions supported by R. All base distri-
butions are included, with the exception of wilcox and signedrank. For more resources please
see the References, and for complete details and formulas see the charfunc vignette, which can
be accessed by vignette("charfunc") at the command prompt. Only the simplest formulas are
listed below.

Usage
cfbeta(t, shape1, shape2, ncp = 0)
cfbinom(t, size, prob)
cfcauchy(t, location = 0, scale = 1)
cfchisq(t, df, ncp = 0)
cfexp(t, rate = 1)
cff(t, df1, df2, ncp, kmax = 10)
cfgamma(t, shape, rate = 1, scale = 1/rate)
cfgeom(t, prob)
cfhyper(t, m, n, k)
cflnorm(t, meanlog = 0, sdlog = 1)
cflogis(t, location = 0, scale = 1)
cfnbinom(t, size, prob, mu)
6 CharFunc

cfnorm(t, mean = 0, sd = 1)
cfpois(t, lambda)
cfsignrank(t, n)
cft(t, df, ncp)
cfunif(t, min=0, max=1)
cfweibull(t, shape, scale = 1)
cfwilcox(t, m, n)

Arguments
t numeric value. Some of the above are vectorized functions.
df degrees of freedom (> 0, maybe non-integer)
df1, df2 numerator and denominator degrees of freedom, must be positive
k the number of balls drawn from the urn.
kmax the number of terms in the series.
lambda vector of (positive) means.
location, scale
location and scale parameters; scale must be positive.
m the number of white balls in the urn.
meanlog, sdlog mean and standard deviation of the distribution on the log scale with default
values of 0 and 1 respectively.
mean vector of means.
min, max (unif) lower and upper limits of the distribution. Must be finite and in the correct
order.
mu (nbinom) alternative parametrization via mean
n the number of black balls in the urn.
ncp non-centrality parameter
prob probability of success in each trial.
rate an alternative way to specify the scale; must be positive.
sd vector of standard deviations.
size number of trials (binom) or target for number of successful trials (nbinom).
shape shape parameter, must be positive (gamma, weibull)
shape1, shape2 positive parameters of the Beta distribution.

Details
The characteristic function of a random variable X is defined by

(t) = EeitX

for all < t < .

Every random variable has a characteristic function, and every characteristic function uniquely
determines the distribution of its associated random variable. For more details on characteristic
functions and their properties, see Lukacs (1970).
CharFunc 7

Value
a complex number in rectangular (cartesian) coordinates.

Beta distribution
For the probability density function, see dbeta.
The characteristic function for central Beta is given by
(t) =1 F1 (; + , it)
where F is the confluent hypergeometric function calculated with kummerM in the fAsianOptions
package.
As of the time of this writing, we must calculate the characteristic function of the noncentral Beta
with numerical integration according to the definition.

Binomial distribution
For the probability mass function, see dbinom.
The characteristic function is given by
(t) = [peit + (1 p)]n

Cauchy Distribution
For the probability density function, see dcauchy.
The characteristic function is given by
(t) = e( it |t|)

Chi-square Distribution
For the probability density function, see dchisq.
The characteristic function is given by
it
exp( 12it )
(t) =
(1 2it)df /2

Exponential Distribution
For the probability density function, see dexp.
This is the special case of gamma when = 1.

F Distribution
For the probability density function, see df.
For the central F we use confluent hypergeometric function of the second kind, also known as
kummerU, from the fAsianOptions package.
For noncentral F we use confluent hypergeometric function of the first kind. See the vignette for
details.
8 CharFunc

Gamma Distribution
For the probability density function, see dgamma.
The characteristic function is given by
(t) = (1 it)( )

Geometric Distribution
For the probability mass function, see dgeom.
This is the special case of negative binomial when r = 1.

Hypergeometric Distribution
For the probability mass function, see dhyper.
The formula for the characteristic function is based on the Gaussian hypergeometric series, calcu-
lated with hypergeo in package hypergeo. It is too complicated to be included here; please see the
vignette.

Logistic Distribution
For the probability density function, see dlogis.
The characteristic function is given by
(t) = t/ sinh(t)

Lognormal Distribution
For the probability density function, see dlnorm.
This characteristic function is uniquely complicated and delicate, but there is a recent numerical
algorithm for computation due to Beaulieu (2008). See the vignette and the References.

Negative Binomial Distribution

For the probability mass function, see dnbinom.
The characteristic function is given by
(t) = (p/(1 (1 p) eit ))r

Normal Distribution
For the probability density function, see dnorm.
The characteristic function is 2
2 /2
(t) = eit+t

Poisson Distribution
For the probability mass function, see dpois.
The characteristic function is it
1)
(t) = e(e
CharFunc 9

Wilcoxon Sign Rank Distribution

For the probability density function, see dsignrank.
The characteristic function is calculated according to the definition.

Students t Distribution
For the probability density function, see dt.
See the vignette for a formula for the characteristic function for central t.
As of the time of this writing, we must calculate the characteristic function of the noncentral t with
numerical integration according to the definition.

Continuous Uniform Distribution

For the probability density function, see dunif.
The characteristic function is
eitb eita
(t) =
(b a)it

Weibull Distribution
For the probability density function, see dweibull.
We must at the time of this writing calculate the characteristic function with numerical integration
according to the definition.

Wilcoxon Rank Sum Distribution

For the probability density function, see dwilcox.
The characteristic function is calculated according to the definition.

Author(s)
G. Jay Kerns <[email protected]>.

Source
For clnorm a fast numerical algorithm is used that originated with and was published and commu-
nicated to me by N. C. Beaulieu: see

References
Abramowitz, M. and Stegun, I. A. (1972) Handbook of Mathematical Functions. New York: Dover.
Beaulieu, N.C. (2008) Fast convenient numerical computation of lognormal characteristic functions,
IEEE Transactions on Communications, Volume 56, Issue 3, 331333.
Hurst, S. (1995) The Characteristic Function of the Student-t Distribution, Financial Mathematics
Research Report No. FMRR006-95, Statistics Research Report No. SRR044-95.
Johnson, N. L., Kotz, S., and Kemp, A. W. (1992) Univariate Discrete Distributions, Second Edi-
tion. New York: Wiley.
10 countrep

Johnson, N. L., Kotz, S. and Balakrishnan, N. (1995) Continuous Univariate Distributions, volume
1. New York: Wiley.
Johnson, N. L., Kotz, S. and Balakrishnan, N. (1995) Continuous Univariate Distributions, volume
2. New York: Wiley.
Lukacs, E. (1970) Characteristic Functions, Second Edition. London: Griffin.

See Also
besselK kummerM kummerU hypergeo

countrep Count Repetitions

Description
Counts the number of repetitions of vals in a given vector x.

Usage
countrep(x, ...)

## Default S3 method:
countrep(x, vals = unique(x), nrep = 2, ...)

## S3 method for class 'data.frame'

countrep(x, ...)

Arguments
x an object in which repeats should be counted.
vals values that may be repeated.
nrep exact number of repeats desired, defaults to pairs.
... further arguments to be passed to or from other methods.

Details
This is a generic function, with methods supplied for data frames and vectors. The default behavior
counts the number of pairs of elements of x. One can find the number of triples, etc., by changing
the nrep argument. If there are specific values for which one is looking for repeats, these can be
specified with the vals argument. Note that the function only checks for exactly nrep instances, so
two pairs of a specific element would be counted as 0 pairs and 1 quadruple. See the examples.
The data frame method uses apply to apply countrep.default to each row of the data frame.

Value
If x is a vector, then the value is an integer. If x is a data frame then the value is a vector, with entries
the corresponding value for the respective rows of x
empirical 11

Author(s)
G. Jay Kerns <[email protected]>.

See Also
isrep

Examples
x <- c(3,3,2,2,3,3,4,4)
countrep(x) # one pair each of 2s and 4s
countrep(x, nrep = 4)
countrep(x, vals = 4) # one pair of 4s

empirical Empirical Summary of a Simulation

Description
Calculates relative frequencies of the rows of a data frame.

Usage
empirical(x)

Arguments
x a data frame.

Details
The function works by adding a probs column to x with equally likely entries of 1/n, where n is the
number of rows. Then it aggregates the duplicated rows of x while accumulating the probabilities
associated with each.

Value
A data frame formed by aggregating the rows of x. A probs column is added giving the relative
frequencies of each of the rows.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
sim
12 euchredeck

Examples
S <- tosscoin(2, makespace = TRUE)
sims <- sim(S, ntrials = 50000)
empirical(sims)

euchredeck A Deck of Playing Cards for Euchre

Description

The title says it all.

Usage

euchredeck(benny= FALSE, makespace = FALSE)

Arguments

benny logical.
makespace logical.

Details

This is a conventional Euchre deck which uses a deck of 24 standard playing cards consisting of
Ace, King, Queen, Jack, 10, and 9 of each of the four suits. If benny = TRUE then a Joker is added
to the deck.

Value

A data frame with columns value and suit, and optionally a column of equally likely probs.

See Also

cards, tosscoin, and roulette

Examples
euchredeck()
euchredeck(benny = TRUE, makespace = TRUE)
gen2wayTable 13

gen2wayTable Generate Two-way Tables

Description
A function to randomly generate arbitrary two-way tables

Usage
gen2wayTable(n = sample(100:500, size = 1), pmatrix = matrix(1:12, nrow = 3),
dmnames = list(X = paste("x", 1:nrow(pmatrix), sep = ""),
Y = paste("y", 1:ncol(pmatrix), sep = "")),
addmargins = TRUE, as.df = FALSE, untable = TRUE)

Arguments
n sum total observations
pmatrix matrix of nonnegative weights for the probability distribution
dmnames names of the table dimensions
addmargins should margins be added?
as.df table will be returned as a data frame
untable should counts be untabled to single observation per row

Value
An object of class table containing the generated values.

Author(s)
G. Jay Kerns

genIndepTable Generate Independent Two-way Table

Description
A function to generate a two-way table with independent margins

Usage
genIndepTable(n = sample(100:500, size = 1), prow = 1:3, pcol = 1:4,
dmnames = list(X = paste("x", 1:length(prow), sep = ""),
Y = paste("y", 1:length(pcol), sep = "")), addmargins = TRUE,
as.df = FALSE, untable = TRUE)
14 genLogRegData

Arguments
n sum total of observations generated
prow nonnegative weights for the row marginal distribution
pcol nonnegative weights for the col marginal distribution
dmnames names for the table dimensions
addmargins should margins be added to the table
as.df should the result be returned as a data frame
untable if true then data frame will be expanded to one observation per row

Details
This function will generate a two-way table with independent marginal distributions.

Value
Either an object of class table or a data frame.

Author(s)
G. Jay Kerns

genLogRegData Generate data for logistic regression

Description
This function generates data ready for a logistic regression model

Usage
genLogRegData(xdata, beta = rep(1, ncol(xdata)), yname = "y")

Arguments
xdata the model matrix
beta vector of parameters to multiply the model matrix
yname the name for the generated y values

Details
This function generates data ready for a logistic regression model

Value
A data frame with the model matrix and the generated y values added
genXdata 15

Author(s)

G. Jay Kerns

genXdata Generate continuous model matrix data

Description

This function generates correlated normal data to serve as a model matrix in a regression model.

Usage

genXdata(n, nvar = 1, mu = rep(0, nvar), Sigma = diag(length(mu)),

varnames = paste("x", 1:length(mu), sep = ""), roundto = NULL)

Arguments

n how many rows

nvar how many columns
mu the mean of the multivariate normal distribution
Sigma the variance-covariance matrix of the normal distribution
varnames how you would like the variables to be named in the result
roundto number of places to round the generated values

Details

This function generates correlated normal data to serve as a model matrix in a regression model.

Value

A data frame of generated data

Author(s)

G. Jay Kerns
16 iidspace

iidspace Independent Identical Experiments

Description
Sets up a probability space corresponding to independent, identical experiments.

Usage
iidspace(x, ntrials, probs = NULL)

Arguments
x a vector of outcomes.
ntrials number of times to perform the experiment.
probs vector of non-negative weights corresponding to x.

Details
The elementary experiment to be repeated consists of drawing an element of x according to the
probabilities contained in probs. The entries of probs need not sum to one, but they will be
normalized before any computations. If probs is not specified, the equally likely model will be
assumed.

Value
A data frame, with a probs column, where probs is calculated to be the probability of observing
the outcome in its row under the assumption of independence and identical distribution of draws
from x.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
probspace

Examples
iidspace( 1:6, ntrials = 3) # same as rolldie(3)
iidspace( 1:6, ntrials = 3, probs = 3:8 ) # unbalanced die
intersect 17

intersect Intersection of Subsets

Description
Calculates the intersection of subsets of a probability space. Comparisons are made row-wise, so
that in the data frame case, intersect(A,B) is a data frame with those rows that are both in A and
in B.

Usage
intersect(x, ...)

## Default S3 method:
intersect(x, y, ...)

## S3 method for class 'data.frame'

intersect(x, y, ...)

## S3 method for class 'ps'

intersect(x, y, ...)

Arguments
x, y vectors, data frames, or ps objects containing a sequence of elements (concep-
tually).
... further arguments to be passed to or from other methods.

Details
This is a generic function, extended from the intersect function in the base package. The ele-
ments of intersect(x,y) are those elements in x and in y. The original definition is preserved in
the case that x and y are vectors of the same mode.

Value
A vector, data frame, or subset of a probability space of the same type as its arguments.

Author(s)
G. Jay Kerns <[email protected]>, based on a suggestion made by Brian Ripley in R-devel, 12/11/07.

See Also
union, setdiff
18 is.probspace

Examples
S <- cards()
A <- subset(S, suit == "Heart")
B <- subset(S, rank == "A" )
intersect(A, B)

is.probspace Testing for a Probability Space

Description
Decides whether a given object is a probability space.

Usage
is.probspace(x)

Arguments
x an object for which probability space status should be checked.

Details
It first checks if the class of the object includes ps, and if so TRUE is returned. If not, then it checks
that the object is a data frame and contains a probs column. Lastly, it checks whether all entries of
probs are nonnegative. Note that it does not check whether the sum of probs is one, to allow for
the possibility that the input object is a proper subset of a probability space.

Value
Logical.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
probspace

Examples
S <- rolldie(3, makespace = TRUE)
is.probspace(S)
isin 19

isin Test Whether One Vector Is In Another Vector

Description

See the title.

Usage

isin(x, ...)

## Default S3 method:
isin(x, y, ordered = FALSE, ...)

## S3 method for class 'data.frame'

isin(x, ...)

Arguments

x, y vectors.
ordered logical.
... further arguments to be passed to or from other methods.

Details

The function will only return TRUE if every element of y is present in the vector x, counting multi-
plicity. See the examples below. Of ordered = TRUE, then elements must be in the vector x in the
order specified in y. Compare this to the behavior of the %in% function in the base package.
This is a generic function with a moethod for data frames, which applies isin() to each row of the
data frame, with a vector as a result.

Value

Logical, or a vector of logicals.

Author(s)

G. Jay Kerns <[email protected]>.

See Also

isrep
20 isrep

Examples
x <- 1:10
y <- 3:7
z <- c(3,3,7)
isin(x,y)
isin(x,z)
isin(x, c(3,4,5), ordered = TRUE)
isin(x, c(3,5,4), ordered = TRUE)

S <- rolldie(4)
subset(S, isin(S, c(2,2,6), ordered = TRUE))

isrep Is Repeated in a Vector

Description
Tests for a certain number of repetitions of vals in a given vector x.

Usage
isrep(x, ...)

## Default S3 method:
isrep(x, vals = unique(x), nrep = 2, ...)

## S3 method for class 'data.frame'

isrep(x, ...)

Arguments
x an object with potential repeated values.
vals values that may be repeated.
nrep exact number of repeats desired, defaults to pairs.
... further arguments to be passed to or from other methods.

Details
This is a generic function, with methods supplied for data frames and vectors. The default behavior
tests for existence of pairs of elements of x. One can test existence of triples, etc., by changing
the nrep argument. If there are specific values for which one is looking for repeats, these can be
specified with the vals argument. Note that the function only checks for exactly nrep instances, so
two pairs of a specific element would be counted as 0 pairs and 1 quadruple. See the examples.
The data frame method uses apply to apply isrep.default to each row of the data frame.
marginal 21

Value
Logical.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
countrep

Examples
x <- c(3,3,2,2,3,3,4,4)
isrep(x) # one pair each of 2s and 4s
isrep(x, nrep = 4)
isrep(x, vals = 4) # one pair of 4s

marginal Marginal Distributions

Description
Computes the marginal distribution of a set of variables.

Usage
marginal(space, vars = NULL)

Arguments
space a data frame probability space or a subset of one.
vars an optional character vector of variable names in space.

Details
If vars is not specified, then marginal() will set vars to be all non-probs columns, which can be
useful in the case that it is desired to aggregate duplicated rows.

Value
A data frame with a probs column.

Author(s)
G. Jay Kerns <[email protected]>.
22 noorder

See Also
See addrv for adding random variables to a data frame probability space.

Examples
S <- rolldie(3, makespace = TRUE)
marginal(S, vars = c("X1", "X2"))

noorder Sort and Merge Probability Space Outcomes

Description
This function sorts the rows (outcomes) of a data frame probability space, effectively removing
the original order present) and aggregates the sorted rows into a new probability data frame with
unique, sorted outcomes.

Usage
noorder( space )

Arguments
space a data frame probability space or a subset of one.

Details
The data frame space must have at least two non-probs columns or an error will result.

Value
A data frame with a probs column and sorted, unique rows.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
addrv, marginal

Examples
S <- tosscoin(3, makespace = TRUE)
noorder(S)
nsamp 23

nsamp Number of Samples from an Urn

Description
Calculates the number of samples from an urn under different sampling scenarios.

Usage
nsamp(n, k, replace = FALSE, ordered = FALSE)

Arguments
n an integer or integer vector.
k an integer or integer vector.
replace logical indicating whether sampling should be done with replacement.
ordered logical indicating whether order among samples is important.

Details
The nsamp() function will calculate the number of samples from an urn under assorted assumptions
on the sampling procedure. The arguments are: n, the number of (distinguishable) objects in the
urn, k, the sample size, and replace, ordered as documented in urnsamples.
nsamp() is vectorized, so that entering vectors instead of numbers for n, k, replace, and ordered
results in a vector of corresponding answers.
The formulas used in the four possible combinations of replace and ordered are as follows.

When replace = TRUE and ordered = TRUE, the value is nk .

When replace = FALSE and ordered = TRUE, the value is n!/(n k)!.
When replace = FALSE and ordered = FALSE, the value is n!/[k!(n k)!].
When replace = TRUE and ordered = FALSE, the value is (n 1 + k)!/[(n 1)!k!].

Value
A number.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
urnsamples
24 permsn

Examples
nsamp(n=3, k=2, replace = TRUE, ordered = TRUE)
nsamp(n=3, k=2, replace = TRUE, ordered = FALSE)
nsamp(n=3, k=2, replace = FALSE, ordered = FALSE)
nsamp(n=3, k=2, replace = FALSE, ordered = TRUE)

permsn Generate All Permutations of x Elements Taken m at a Time

Description

Generate all permutations of the elements of x taken m at a time. If x is a positive integer, returns
all permutations of the elements of seq(x) taken m at a time.

Usage

permsn(x, m)

Arguments

x vector source for permutations, or integer n for x <- seq(n).

m number of elements to permute.

Value

a list or array (in nondegenerate cases).

Author(s)

G. Jay Kerns <[email protected]>, modified from the combn function in the package utils.

See Also

combn

Examples
permsn(3,2)
prob 25

prob Probability and Conditional Probability

Description
Calculates probability and conditional probability of events.

Usage
Prob(x, ...)

## Default S3 method:
Prob(x, event = NULL, given = NULL, ...)

## S3 method for class 'ps'

Prob(x, event = NULL, given = NULL, ...)

Arguments
x a probability space or a subset of one.
event logical expression indicating elements or rows of space to keep: missing values
are taken as false.
given either a subset of a probability space or a logical expression indicating elements
or rows of space to keep: missing values are taken as false.
... further arguments to be passed to or from other methods.

Details
This function calculates the probability of events or subsets of a given sample space. Conditional
probability is also implemented. In essence, the Prob() function operates by summing the probs
column of its argument. It will find subsets on the fly if desired.
The event argument is used to define a subset of x, that is, the only outcomes used in the probability
calculation will be those that are elements of x and satisfy event simultaneously. In other words,
Prob(x,event) calculates Prob(intersect(x, subset(x, event))). Consequently, x should
be the entire probability space in the case that event is non-null.
There is some flexibility in the given argument in that it can be either a data frame or it can be a
logical expression that defines the subset. However, that flexibility is limited. In particular, if given
is a logical expression, then event must also be specified (also a logical expression). And in this
case, the argument x should be the entire sample space, not a subset thereof.

Value
A number in the interval [0,1].

Author(s)
G. Jay Kerns <[email protected]>.
26 probspace

See Also
probspace, iidspace

Examples
S <- rolldie(times = 3, makespace = TRUE )
Prob(S, X1+X2 > 9 )
Prob(S, X1+X2 > 9, given = X1+X2+X3 > 7 )

probspace Probability Spaces

Description
Forms a probability space from a set of outcomes and (optional) vector of probabilities.

Usage
probspace(x, ...)

## Default S3 method:
probspace(x, probs, ...)

## S3 method for class 'list'

probspace(x, probs, ...)

Arguments
x a vector, data frame, or list of outcomes.
probs a vector of non-negative weights of the same length as outcomes
... further arguments to be passed to or from other methods.

Details
The elements of probs will be normalized to ensure that their sum is one. If probs is not specified,
then the equally likely model is assumed in which every outcome has the same probability.

Value
If outcomes is a vector or data frame, then the value is a data frame with an added probs column.
If outcomes is a list, then the value is a list with components outcomes (the supplied list) and a
probs component.

Author(s)
G. Jay Kerns <[email protected]>.
rolldie 27

See Also
iidspace

Examples
R <- rolldie(3)
probspace(R)

rolldie Rolling a Die

Description
Sets up a sample space for the experiment of rolling a die repeatedly.

Usage
rolldie(times, nsides = 6, makespace = FALSE)

Arguments
times number of rolls.
nsides number of sides on the die.
makespace logical.

Details
The function uses expand.grid() to generate all possible rolls resulting from the experiment of
rolling a die. Sides on the die are 1:nsides. Columns of the data frame are called X1, X2, up to
Xtimes

Value
A data frame, with an equally likely probs column if makespace is TRUE.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
tosscoin

Examples
rolldie(2)
rolldie(3, nsides = 4)
rolldie(3, nsides = 4, makespace = TRUE)
28 roulette

roulette Roulette

Description

Sets up a sample space for the experiment of spinning a roulette wheel once.

Usage

roulette(european = FALSE, makespace = FALSE)

Arguments

european logical.
makespace logical.

Details

If european is TRUE, then a traditional EU roulette wheel with 37 pockets is used, otherwise, a
standard US roulette wheel with 38 pockets is used. Entries in the data frame are ordered in the
customary way to facilitate the calculation probabilities regarding called bets.

Value

A data frame, with columns num and color, and an equally likely probs column if makespace is
TRUE.

Author(s)

G. Jay Kerns <[email protected]>.

See Also

cards

Examples
roulette()
roulette(european = TRUE, makespace = TRUE)
setdiff 29

setdiff Set Difference of Subsets

Description

Calculates the (nonsymmetric) set difference of subsets of a probability space.

Usage

setdiff(x, ...)

## Default S3 method:
setdiff(x, y, ...)

## S3 method for class 'data.frame'

setdiff(x, y, ...)

## S3 method for class 'ps'

setdiff(x, y, ...)

Arguments

x, y vectors, data frames, or ps objects containing a sequence of items (conceptu-

ally).
... further arguments to be passed to or from other methods.

Details

This function operates row-wise on dataframes, and element-wise among the outcomes of ps ob-
jects. The elements of setdiff(x,y) are those elements in x but not in y. The definition is taken
to match the version in the base package.

Value

A data frame or subset of a probability space of the same type as its arguments.

Author(s)

G. Jay Kerns <[email protected]>, essentially verbatim from a suggestion made by Brian Ripley on
R-devel, 12/11/07.

See Also

intersect, union
30 sim

Examples
S <- cards()
A <- subset(S, suit == "Heart")
B <- subset(S, rank == "A" )
setdiff(B, A)

sim Simulate Draws from a Sample Space

Description
Simulates the experiment of drawing from a sample space.

Usage
sim(x, ...)

## Default S3 method:
sim(x, ntrials, ...)

## S3 method for class 'ps'

sim(x, ntrials, ...)

Arguments
x a probability space or a subset of one.
ntrials number of times to repeat the experiment.
... further arguments to be passed to or from other methods.

Details
The sim() function is a wrapper for sample(), except that it strips the probs component from
the result and (if x is a data frame) renames the rownames of the data frame consecutively from
1:ntrials.

Value
A data frame if space is a data frame, or a list if space is of class ps.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
empirical
subset 31

Examples
S <- cards(makespace = TRUE)
sim(S, ntrials = 5)

T <- urnsamples(S, 2)
U <- probspace(T)
sim(U, ntrials = 4)

subset Subsets of Probability Spaces

Description
This is a method for subset() for the case when the input object is a probability space of class ps.

Usage
subset(x, ...)

## S3 method for class 'ps'

subset(x, subset, ...)

Arguments
x a probability space.
subset logical expression indicating elements or rows of space to keep: missing values
are taken as false.
... further arguments to be passed to or from other methods.

Details
This function simply extends the existing subset() function to ps objects.

Value
A ps object, a subset of a probability space.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
intersect, setdiff, union, isin
32 tosscoin

Examples
L <- tosscoin(2)
M <- urnsamples(L, 3)
N <- probspace(M)
subset(N, all(toss1=="H"))
subset(N, any(toss2=="T"))

tosscoin Tossing a Coin

Description
Sets up a sample space for the experiment of tossing a coin repeatedly with the outcomes "H" or
"T".

Usage
tosscoin( times, makespace = FALSE )

Arguments
times number of tosses.
makespace logical.

Details
The function uses expand.grid() to generate all possible sequences of flips resulting from the ex-
periment of tossing a coin. Columns of the dataframe are denoted toss1, toss2, up to tosstimes,

Value
A data frame, with an equally likely probs column if makespace is TRUE.

Author(s)
G. Jay Kerns <[email protected]>.

See Also
rolldie

Examples
tosscoin(2)
tosscoin(3, makespace = TRUE)
union 33

union Union of Subsets

Description

Calculates the union of subsets of a probability space.

Usage

union(x, ...)

## Default S3 method:
union(x, y, ...)

## S3 method for class 'data.frame'

union(x, y, ...)

## S3 method for class 'ps'

union(x, y, ...)

Arguments

x, y vectors, data frames, or ps objects containing a sequence of items (conceptually)

... further arguments to be passed to or from other methods.

Details

This function operates row-wise on dataframes, and element-wise among the outcomes of ps ob-
jects. The elements of union(x,y) are those elements in x or y, or both. The definition is taken to
match the version in the base package.

Value

A data frame or subset of a probability space of the same type as its arguments.

Author(s)

G. Jay Kerns <[email protected]>, based on a suggestion made by Brian Ripley in R-devel, 12/11/07.

See Also

intersect, setdiff
34 urnsamples

Examples
S <- cards()
A <- subset(S, suit == "Heart")
B <- subset(S, rank == "A" )
union(A, B)

urnsamples Sampling from Urns

Description
This function creates a sample space associated with the experiment of sampling distinguishable
objects from an urn.

Usage
urnsamples(x, ...)

## Default S3 method:
urnsamples(x, size, replace = FALSE, ordered = FALSE, ...)

## S3 method for class 'data.frame'

urnsamples(x, size, replace = FALSE, ordered = FALSE, ...)

Arguments
x a vector or data frame from which sampling should take place.
size number indicating the sample size.
replace logical indicating whether sampling should be done with replacement.
ordered logical indicating whether order among samples is important.
... further arguments to be passed to or from other methods.

Details
The function operates on the indices of the urn (or rows, in the case urn is a data frame). It then
takes those samples and substitutes back into urn to generate the entries of the data frame (or list,
respectively). In the case that urn has repeated values, the result will be repeated values in the
output.
Note that urnsamples strips x of any existing probs column before sampling.

Value
A data frame if urn is a vector, and a list if urn is a data frame.
urnsamples 35

Author(s)
G. Jay Kerns <[email protected]>.

See Also
iidspace, probspace nsamp

Examples
urnsamples(1:10, size = 5)
S <- cards()
urnsamples(S, size = 2)
Index

Topic \textasciitildedatagen besselK, 10

genLogRegData, 14
genXdata, 15 cards, 4, 12, 28
Topic datagen cfbeta (CharFunc), 5
gen2wayTable, 13 cfbinom (CharFunc), 5
genIndepTable, 13 cfcauchy (CharFunc), 5
Topic data cfchisq (CharFunc), 5
cards, 4 cfexp (CharFunc), 5
euchredeck, 12 cff (CharFunc), 5
Topic distribution cfgamma (CharFunc), 5
CharFunc, 5 cfgeom (CharFunc), 5
Topic manip cfhyper (CharFunc), 5
addrv, 3 cflnorm (CharFunc), 5
empirical, 11 cflogis (CharFunc), 5
marginal, 21 cfnbinom (CharFunc), 5
Topic methods cfnorm (CharFunc), 5
countrep, 10 cfpois (CharFunc), 5
Topic misc cfsignrank (CharFunc), 5
iidspace, 16 cft (CharFunc), 5
intersect, 17 cfunif (CharFunc), 5
is.probspace, 18 cfweibull (CharFunc), 5
isin, 19 cfwilcox (CharFunc), 5
isrep, 20 CharFunc, 5
noorder, 22 combn, 24
nsamp, 23 countrep, 10, 21
permsn, 24
prob, 25 dbeta, 7
probspace, 26 dbinom, 7
rolldie, 27 dcauchy, 7
roulette, 28 dchisq, 7
setdiff, 29 dexp, 7
sim, 30 df, 7
subset, 31 dgamma, 8
tosscoin, 32 dgeom, 8
union, 33 dhyper, 8
urnsamples, 34 dlnorm, 8
Topic package dlogis, 8
prob-package, 2 dnbinom, 8
dnorm, 8
addrv, 3, 22 dpois, 8

36
INDEX 37

dsignrank, 9
dt, 9
dunif, 9
dweibull, 9
dwilcox, 9

empirical, 11, 30
euchredeck, 12

gen2wayTable, 13
genIndepTable, 13
genLogRegData, 14
genXdata, 15

hypergeo, 8, 10

iidspace, 16, 26, 27, 35

intersect, 17, 29, 31, 33
is.probspace, 18
isin, 19, 31
isrep, 11, 19, 20

kummerM, 7, 10
kummerU, 7, 10

marginal, 21, 22

noorder, 22
nsamp, 23, 35

permsn, 24
Prob (prob), 25
prob, 25
prob-package, 2
Prob.default (prob), 25
Prob.ps (prob), 25
probspace, 16, 18, 26, 26, 35

rolldie, 5, 27, 32
roulette, 5, 12, 28

setdiff, 17, 29, 31, 33

sim, 11, 30
subset, 31

tosscoin, 5, 12, 27, 32

transform, 4

union, 17, 29, 31, 33

urnsamples, 23, 34