| Title: | Symbolic Central and Noncentral Moments of the Multivariate Normal Distribution |
|---|---|
| Description: | Symbolic central and non-central moments of the multivariate normal distribution. Computes a standard representation, LateX code, and values at specified mean and covariance matrices. |
| Authors: | Kem Phillips |
| Maintainer: | Kem Phillips <[email protected]> |
| License: | GPL |
| Version: | 1.2.1 |
| Built: | 2024-12-23 06:45:25 UTC |
| Source: | CRAN |
Generic method for class moment to compute the numerical value of a moment at a specified covariance matrix from the output of callmultmoments
## S3 method for class 'moment' evaluate(object,sigma)## S3 method for class 'moment' evaluate(object,sigma)
object |
an object of class 'moment' |
sigma |
an upper-triangular matrix of covariance terms expressed as a vector at which the moment is to be evaluated |
object is normally the output of a call to callmultmoment. This is a list with first component the moment itself, the second component the set of upper-triangular matrices representing the moment, and the third component containing their corresponding coefficients. This is an object of class 'moment'.
numeric value of the moment at the specified covariance matrix
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
callmultmoments and the simulate and toLatex methods from the symmoments package
evaluate(callmultmoments(c(1,2,3,4)),c(4,2,1,1,3,1,1,2,1,2)) # evaluates the moment at c(1,2,3,4) at the following covariance matrix # 4 2 1 1 # 2 3 1 1 # 1 1 2 1 # 1 1 1 2evaluate(callmultmoments(c(1,2,3,4)),c(4,2,1,1,3,1,1,2,1,2)) # evaluates the moment at c(1,2,3,4) at the following covariance matrix # 4 2 1 1 # 2 3 1 1 # 1 1 2 1 # 1 1 1 2
Evaluate the expected value of a multivariate polynomial assuming a specified non-central multivariate distribution.
evaluate_expected.polynomial(poly,mu,sigma, envir='symmoments')evaluate_expected.polynomial(poly,mu,sigma, envir='symmoments')
poly |
either an object of class 'mpoly' or 'multipol', or a list with components for coefficients and powers. |
mu |
a vector of real numbers representing the mean of the multivariate distribution |
sigma |
an vector giving an upper-triangular matrix representing the covariance matrix of the multivariate distribution |
envir |
a character variable specifying the environment containing the central moments needed for the calculation |
This function looks in the environment specified in the envir argument for the central moments needed in the calculation. The default is the symmoments environment. The computation stops with an error message if a required moment is not found in envir.
expected value of the multivariate polynomial at the specified multivariate normal mean and covariance matrix
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
See the evaluate_noncentral and make.all.moments functions.
# define a mpoly object for a multivariate polynomial and determine # its expected value at specified mean and covariance matrix: # note that all moments up to c(2,3,2) must exist in the symmoments # environment. Use make.all.moments(c(2,3,2)) if necessary. # use library(mpoly) for first statement below. # t0 <- mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3),c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # evaluate_expected.polynomial(t0,c(1,2,3),c(1,0,0,1,0,1))# define a mpoly object for a multivariate polynomial and determine # its expected value at specified mean and covariance matrix: # note that all moments up to c(2,3,2) must exist in the symmoments # environment. Use make.all.moments(c(2,3,2)) if necessary. # use library(mpoly) for first statement below. # t0 <- mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3),c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # evaluate_expected.polynomial(t0,c(1,2,3),c(1,0,0,1,0,1))
Computes the numerical value of a non-central moment at a specfied mean and specified covariance matrix
evaluate_noncentral(moment,mu,sigma,envir='symmoments')evaluate_noncentral(moment,mu,sigma,envir='symmoments')
moment |
a vector of non-negative integers representing the non-central moment to be evaluated |
mu |
a vector of real numbers representing the mean of the multivariate normal distribution |
sigma |
an upper-triangular matrix of covariance terms for the multivariate normal distribution expressed as a vector at which the moment is to be evaluated |
envir |
a character variable specifying the environment containing the central moments needed for the calculation |
This function looks in the environment specified in the envir argument for the central moments needed in the calculation. The default is the symmoments environment. All even moments less than or equal to the moment argument are required. The computation stops with an error message if a required moment is not found in envir.
numeric value of the moment at the specified mean and covariance matrix
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
See the evaluate.moment and make.all.moments functions.
# evaluate_noncentral(c(3,1,2),c(3,4,1),c(4,2,1,3,1,2)) # evaluates the expected value of X1**3 X2 X3**2 at mean c(1,2,3) # and at the following covariance matrix # 4 2 1 # 2 3 1 # 1 1 2 # requires all moments up to c(3,1,2) to exist in the symmoments environment. # use make.all.moments(c(3,1,2)) if necessary. # use moments in the global environment: # evaluate_noncentral(c(3,1,2),c(3,4,1),c(4,2,1,3,1,2),'.GlobalEnv')# evaluate_noncentral(c(3,1,2),c(3,4,1),c(4,2,1,3,1,2)) # evaluates the expected value of X1**3 X2 X3**2 at mean c(1,2,3) # and at the following covariance matrix # 4 2 1 # 2 3 1 # 1 1 2 # requires all moments up to c(3,1,2) to exist in the symmoments environment. # use make.all.moments(c(3,1,2)) if necessary. # use moments in the global environment: # evaluate_noncentral(c(3,1,2),c(3,4,1),c(4,2,1,3,1,2),'.GlobalEnv')
Called by callmultmoments to compute representation of a multivariate normal moment using recursive algorithm
multmoments(moment,current.matrix,current.cell,moment.rep,row_col)multmoments(moment,current.matrix,current.cell,moment.rep,row_col)
moment |
vector c(k1,...,kn) specifying the moment X1**k1 *...* Xn**kn |
current.matrix |
upper-triangular integer matrix under consideration in recursion |
current.cell |
cell in current matrix under consideration in recursion |
moment.rep |
current set of representations; mult.moments adds each satisfying matrix to moment.rep |
row_col |
matrix giving rows and columns for square matrix for each cell |
Each row of the representation gives the exponents for a single product of covariance terms. For example, (1,2,0) represents S11**1 S12**2 S22**0, where the Sij are the covariances.
This function would normally only be called by callmultmoments.
moment representation, moment.rep, augmented with additional representations
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
callmultmoments (symmoments)
Symbolically computes and numerically evaluates multivariate normal moments E[X1**2 *...* Xn**n], where (X1,...,Xn) ~ N(mu,S), in terms of mu and S elements.
Produces Latex code for the moment.
Computes numerical moments at specified means and covariance matrices.
Also converts between moment L-matrices, phylo objects, and matching objects.
| Package: | symmoments |
| Type: | Package |
| Version: | 1.2 |
| Date: | 2018-05-30 |
| License: | GPL 2 |
| LazyLoad: | yes |
A representation of a central moment of the multivariate normal distribution, given by a
positive integer vector c(k1,k2,...,kn), is obtained from the function callmultmoments.
This function initializes variables and calls the function multmoments which determines
a representation of a multivariate moment using a recursive algorithm.
The representation is given class 'moment'.
The print method prints the representation of a multivariate moment.
The toLatex method uses the output of callmultmoments to determine the LaTeX code
for the moment sorted lexicographically.
The generic evaluate method uses the output of callmultmoments to determine
the value of the moment for a specified covariance matrix.
The simulate method is used to approximate a (possibly non-central) moment
using Monte Carlo integration.
The following functions compute non-central moments and do related computations:
The toLatex_noncentral function computes the Latex representations of a non-central moment.
The evaluate_noncentral computes the value of a non-central moment.
The evaluate_expected.polynomial function evaluates the expected value
of a multivariate polynomial defined by a list, multipol object, or mpoly object.
The convert.multipol function converts between multipol objects and
multivariate polynomials defined by lists.
The convert.mpoly function converts between mpoly objects and
multivariate polynomials defined by lists.
The tounsorted function converts a sorted moment (e.g. m123) to an unsorted moment (e.g. m312).
The make.all.moments function computes all moments up to a specified size and places them in the symmoments environment.
The integrate.polynomial function integrates a multivariate polynomial against
the normal distribution using ordinary integration.
The functions toMoment, toNewick, and toMatching
convert among moment L-matrices, Newick trees, and ape matching objects.
The mvtnorm package must be loaded for the simulate method. The cubature package must be loaded for the integrate.polynomial function. The combinat package must be loaded for the toMoment function.
Maintainer: Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
# Compute the moment for the 4-dimensional moment c(1,2,3,4): callmultmoments(c(1,2,3,4)) # Print the representation of the 4-dimensional moment c(1,2,3,4): print(callmultmoments(c(1,2,3,4))) # Compute the LaTeX representation of the central moment c(1,2,3,4): toLatex(callmultmoments(c(1,2,3,4))) # Write the LaTeX representation to a file using the standard R function (not run): # writeLines(callmultmoments(c(1,2,3,4))),con="yourfilename", sep = "\n") # evaluate the moment c(1,2,3,4) at the following variance-covariance matrix # 4 2 1 1 # 2 3 1 1 # 1 1 2 1 evaluate(callmultmoments(c(1,2,3,4)),c(4,2,1,1,3,1,1,2,1,2)) # Using 10000 samples, estimate the central moment for c(2,4) at the covariance matrix (not run) # 2 1 # 1 4 # and mean (0,0) library(mvtnorm) simulate(callmultmoments(c(2,4)),10000,NULL,c(0,0),c(2,1,1,4)) # Compute Latex representation of a non-central moment # as.matrix(toLatex_noncentral(c(1,3))) # Create all 2-dimensional moment objects with exponents up to 3 # First create the symmoments environment if it does not exist # symmoments <- new.env() # make.all.moments(c(3,3)) # Evaluate a non-central moment at a specified mean and covariance matrix # Note that this invocation requires moments of order up to c(1,3) # to exist in environment symmoments. # evaluate_noncentral(c(1,3),c(1,2),c(1,0,1)) # Create an mpoly object library(mpoly) t0 <- mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3), c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # Convert an mpolyobject to a moment object t1 <<- convert.mpoly(t0) # Convert a moment object to a multipol object t2 <<- convert.multipol(t1) # Convert from multipol back to mpoly through moment mpoly(convert.mpoly(convert.multipol(t2))) # Evaluate the expected value of a multivariate polynomial # Required moments must exist in environment symmoments. # evaluate_expected.polynomial(t0,c(1,2,3),c(1,0,0,1,0,1)) # Create a Newick representation of a tree exam.Newick <- "(((a,b),c),d);" # Convert to phylo format library(ape) exam.phylo <- read.tree(text=exam.Newick) # Convert to matching format exam.matching <- as.matching(exam.phylo) # Convert to L-matrix format exam.L.matrix <- toMoment(exam.matching)# Compute the moment for the 4-dimensional moment c(1,2,3,4): callmultmoments(c(1,2,3,4)) # Print the representation of the 4-dimensional moment c(1,2,3,4): print(callmultmoments(c(1,2,3,4))) # Compute the LaTeX representation of the central moment c(1,2,3,4): toLatex(callmultmoments(c(1,2,3,4))) # Write the LaTeX representation to a file using the standard R function (not run): # writeLines(callmultmoments(c(1,2,3,4))),con="yourfilename", sep = "\n") # evaluate the moment c(1,2,3,4) at the following variance-covariance matrix # 4 2 1 1 # 2 3 1 1 # 1 1 2 1 evaluate(callmultmoments(c(1,2,3,4)),c(4,2,1,1,3,1,1,2,1,2)) # Using 10000 samples, estimate the central moment for c(2,4) at the covariance matrix (not run) # 2 1 # 1 4 # and mean (0,0) library(mvtnorm) simulate(callmultmoments(c(2,4)),10000,NULL,c(0,0),c(2,1,1,4)) # Compute Latex representation of a non-central moment # as.matrix(toLatex_noncentral(c(1,3))) # Create all 2-dimensional moment objects with exponents up to 3 # First create the symmoments environment if it does not exist # symmoments <- new.env() # make.all.moments(c(3,3)) # Evaluate a non-central moment at a specified mean and covariance matrix # Note that this invocation requires moments of order up to c(1,3) # to exist in environment symmoments. # evaluate_noncentral(c(1,3),c(1,2),c(1,0,1)) # Create an mpoly object library(mpoly) t0 <- mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3), c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # Convert an mpolyobject to a moment object t1 <<- convert.mpoly(t0) # Convert a moment object to a multipol object t2 <<- convert.multipol(t1) # Convert from multipol back to mpoly through moment mpoly(convert.mpoly(convert.multipol(t2))) # Evaluate the expected value of a multivariate polynomial # Required moments must exist in environment symmoments. # evaluate_expected.polynomial(t0,c(1,2,3),c(1,0,0,1,0,1)) # Create a Newick representation of a tree exam.Newick <- "(((a,b),c),d);" # Convert to phylo format library(ape) exam.phylo <- read.tree(text=exam.Newick) # Convert to matching format exam.matching <- as.matching(exam.phylo) # Convert to L-matrix format exam.L.matrix <- toMoment(exam.matching)
Computes a multivariate normal moment by initializing variables, calling multmoments, and constructing output
callmultmoments(moment)callmultmoments(moment)
moment |
vector c(k1,... ,kn) specifying the moment X1**k1 *...* Xn**kn |
Each row of the representation gives the exponents for a single product of covariance terms. For example, (1,2,0) represents S11**1 S12** S22**0, where the Sij are the covariances. The full moment is the sum of these terms multiplied by their respective coefficients. If the sum of the exponents is odd, the moment is 0.
A object of class 'moment', which is a list with three components:
moment |
the input moment vector |
representation |
a matrix containing the representation in terms of upper-triangular matrices |
coefficients |
the coefficients corresponding to the rows of the representation |
If the sum of the exponents is odd, returns -1 and prints "Sum of powers is odd. Moment is 0."
If any exponent is negative, returns -2 and prints "All components of the moment must be non-negative."
If any exponent is not an integer, returns -3 and prints "All components of the moment must be integers."
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
multmoments and the methods toLatex, evaluate, and simulate in symmoments
# Compute the moment for the 4-dimensional moment c(1,2,3,4): m.1234 <- callmultmoments(c(1,2,3,4))# Compute the moment for the 4-dimensional moment c(1,2,3,4): m.1234 <- callmultmoments(c(1,2,3,4))
Converts an mpoly object to a simple list representing a multivariate polynomial or a simple list to an mpoly object
convert.mpoly(poly)convert.mpoly(poly)
poly |
an mpoly object or a list giving powers and coefficients defining the polynomial |
The list representation consists of 2 components: 'powers' is a matrix with each row representing the powers of X in one term of the multivariate polynomial. 'coeff' is a vector with each element being the coefficient of the corresponding term in powers
if poly is of class 'mpoly', it is a list with two components shown below. If poly is such a list, the value is the corresponding mpoly object
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
convert.multipol, evaluate.expected.polynomial, integrate.polynomial
# create an mpoly object here (requires library mpoly) (not run) # convert from mpoly to list representation # t0 <- mpoly::mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3), # c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # t1 <- convert.mpoly(t0) # convert from list representation back to an mpoly object # t2 <- convert.mpoly(t1)# create an mpoly object here (requires library mpoly) (not run) # convert from mpoly to list representation # t0 <- mpoly::mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3), # c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # t1 <- convert.mpoly(t0) # convert from list representation back to an mpoly object # t2 <- convert.mpoly(t1)
Converts an multipol object to a simple list representing a multivariate polynomial or a simple list to an multipol object
convert.multipol(poly)convert.multipol(poly)
poly |
a multipol object or a list giving powers and coefficients defining the polynomial |
The list representation consists of 2 components: 'powers' is a matrix with each row representing the powers of X in one term of the multivariate polynomial. 'coeff' is a vector with each element being the coefficient of the corresponding term in powers
if poly is of class 'multipol', it is a list with two components described below. If poly is such a list, the value is the corresponding multipol object
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
convert.multipol, evaluate.expected.polynomial, integrate.polynomial
# create an mpoly object to work with (requires library mpoly) (not run) # t0 <- mpoly::mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3), # c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # convert from mpoly to list representation (not run) # t1 <- convert.mpoly(t0) # convert from list representation to a multipol object(not run) # t2 <- convert.multipol(t1) # convert back to a list representation (not run) # t3 <- convert.multipol(t2)# create an mpoly object to work with (requires library mpoly) (not run) # t0 <- mpoly::mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3), # c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # convert from mpoly to list representation (not run) # t1 <- convert.mpoly(t0) # convert from list representation to a multipol object(not run) # t2 <- convert.multipol(t1) # convert back to a list representation (not run) # t3 <- convert.multipol(t2)
Integrates a multivariate polynomial against a specified non-central multivariate distribution using ordinary integration by invoking the adaptIntegrate function from the cubature package.
integrate.polynomial(poly,mu,sigma,lower=NULL,upper=NULL)integrate.polynomial(poly,mu,sigma,lower=NULL,upper=NULL)
poly |
either an object of class 'mpoly' or 'multipol', or a list with two components for coefficients and powers. |
mu |
a vector giving the mean of the multivariate distribution |
sigma |
a square matrix giving the covariance matrix of the multivariate distribution |
lower |
vectors of the lower limits of integration, one element for each dimension of the moment |
upper |
vectors of the upper limits of integration, one element for each dimension of the moment |
Defaults for lower and upper are -/+ 6 times the standard deviations (square roots of diagonal elements of the covariance matrix). If the polynomial is defined by a list, it has two components, coeff and powers. powers is a matrix. Each row represents the powers for a term in the polynomial. coeff is a vector. Each element is the coefficient of the corresponding power. Example corresponding to example below: list(coeff=c(3,2,-4,1),powers=matrix(c(2,0,0,1,3,0,0,0,2,1,2,1),ncol=3,byrow=TRUE))
the expected value of the polynomial integrated against the multivariate normal distribution
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
evaluate.expected.polynomial, multmoments, evaluate, and simulate in symmoments
# define a mpoly object for a multivariate polynomial, and # determine its expected value at specified mean and covariance matrix: # t0 <- mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3),c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # integrate.polynomial(t0,c(1,2,3),matrix(c(1,0,0,0,1,0,0,0,1),nrow=3,byrow=TRUE))# define a mpoly object for a multivariate polynomial, and # determine its expected value at specified mean and covariance matrix: # t0 <- mpoly(list(c(coef=3,x1=2),c(coef=2,x1=1,x2=3),c(coef=-4,z=2),c(coef=1,x1=1,x2=2,z=1))) # integrate.polynomial(t0,c(1,2,3),matrix(c(1,0,0,0,1,0,0,0,1),nrow=3,byrow=TRUE))
Create all central moment objects of a specified or smaller size in environment symmoments
make.all.moments(moment,verbose=TRUE)make.all.moments(moment,verbose=TRUE)
moment |
vector c(k1,... ,kn) specifying the highest moment to compute |
verbose |
if TRUE (default), the names of the moments are shown as the algorithm progresses; if FALSE, progress is not shown |
Unsorted moments, those with exponents are not in numeric order,
are created in the symmoments environment using the tounsorted
function to transform from the sorted moment. If symmoments does not
exist, the user is prompted to create it using symmoments <- new.env().
If the sorted moment does not exist, it is created.
Moments of lower dimension are not created; for example, if c(2,4) is input, m20 is created, but m2 is not.
Moments are named mij..l, e.g., m136. If any exponent is greater than 9, lower case letters and then upper case letters are used. For example, m3bA is the name of the moment c(3,11,36).
The largest exponent allowed by this scheme is 9+26+26=61,
If an object with a name of this form exists but is not an object of class "moment", it is replaced (overwritten) by the moment object.
all objects of class 'moment' up to the value given in moment are created in environment symmoments
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
callmultmoments, tounsorted
# Compute all moments up to c(3,3) # First create the symmoments environment if it does not exist # symmoments <- new.env() # make.all.moments(c(3,3))# Compute all moments up to c(3,3) # First create the symmoments environment if it does not exist # symmoments <- new.env() # make.all.moments(c(3,3))
Prints an object of class 'moment'
## S3 method for class 'moment' print(x,...)## S3 method for class 'moment' print(x,...)
x |
an object of class 'moment', usually the output of callmultmoments |
... |
Included only for consistency with generic function |
Prints the moment as E[X1**k1 X2**k2 ...]: followed by the lines of the representation with the corresponding coefficient attached
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
callmultmoments (symmoments)
print(callmultmoments(c(1,2,3)))print(callmultmoments(c(1,2,3)))
Computes a multivariate normal moment by Monte Carlo integration
## S3 method for class 'moment' simulate(object,nsim,seed,Mean,Sigma,...)## S3 method for class 'moment' simulate(object,nsim,seed,Mean,Sigma,...)
object |
object of class 'moment' representing E[X1**k1,...,Xn**kn] |
nsim |
the number of samples to generate in computing the integral |
seed |
integer for random number generator (set.seed) |
Mean |
the mean of (X1,...,Xn) |
Sigma |
covariance of (X1**k1,...,Xn**kn), dimension nXn, expressed as a vector by row |
... |
Included only for consistency with generic function |
Approximate value of the moment
Non-central moments can be approximated by specifying Mean. For central moments, set Mean to a vector of 0s.
The mvtnorm package must be loaded for the function rmvnorm.
Kem Phillips <[email protected]>
Rizzo ML (2008). Statistical Computing with R. Chapman & Hall/CRC
callmultmoments and the methods toLatex and evaluate from symmoments
# Using 10000 samples, estimate the central moment for the moment c(2,4) at the covariance matrix # 2 1 # 1 4 # and mean (0,0) library(mvtnorm) simulate(callmultmoments(c(2,4)),10000,NULL,c(0,0),c(2,1,1,4))# Using 10000 samples, estimate the central moment for the moment c(2,4) at the covariance matrix # 2 1 # 1 4 # and mean (0,0) library(mvtnorm) simulate(callmultmoments(c(2,4)),10000,NULL,c(0,0),c(2,1,1,4))
Compute a Latex expression for a noncentral moment
toLatex_noncentral(moment,envir='symmoments')toLatex_noncentral(moment,envir='symmoments')
moment |
vector c(k1,... ,kn) specifying the moment X1**k1 *...* Xn**kn |
envir |
character variable specifying the environment that contains the required central moments |
All required moment objects must exist in the specified environment, with default 'symmoments'. However, if the sorted version of an unsorted moment exists, the tounsorted function is used to obtain it.
A text value giving the Latex representation of moment where X is multivariate normal
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
make.all.moments, tounsorted, callmultmoments and the method toLatex
# Compute the Latex representation of the 2-dimensional moment c(1,3) (not run). # This requires that all moments up to c(1,3) exist in the symmoments environment. # toLatex_noncentral(c(1,3))# Compute the Latex representation of the 2-dimensional moment c(1,3) (not run). # This requires that all moments up to c(1,3) exist in the symmoments environment. # toLatex_noncentral(c(1,3))
Computes a LaTeX representation sorted lexicographically of an object of class 'moment'
## S3 method for class 'moment' toLatex(object,...)## S3 method for class 'moment' toLatex(object,...)
object |
an object of class 'moment', usually the output of callmultmoments |
... |
Included only for consistency with generic function |
The first element of the result is the moment expressed as an expected value (E[...] =). The remaining lines are the LaTex representation broken at appropriate intervals for printing. (Individual terms for high dimensions will still overrun a printed line.) Double backslashes are inserted where LaTeX requires a backslash. These can be reset to single backslashes by writing the output to a file using the R function writeLines from the base package.
Character vector giving the LaTeX code for the symbolic moment
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
callmultmoments and the evaluate method (symmoments)
toLatex(callmultmoments(c(1,2,3)))toLatex(callmultmoments(c(1,2,3)))
Function converts a tree in moment format to matching format.
The input can be an L-matrix object,
a square L matrix,
or an L matrix in reduced upper-triangular (vector) form.
The toMatching function sets its list output to class L-matching,
which has 5 components,
including the tree in matching format.
toMatching(L, type = NULL, tip.label = NULL)toMatching(L, type = NULL, tip.label = NULL)
L |
An |
type |
If object is not of class "L-matrix" and is a square L matrix, then type should be "square'. If it is an L matrix in upper triangular form, type should be "ut". |
tip.label |
Character vector containing labels for tips. If null, labels default to "a"-"z" it at most 26; otherwise, 3-letter labels of the form "aaa", "aab",... |
An L-matrix object is a list with 5 components: "L" is the L-matrix in square form. "L.ut" is the L-matrix in upper-triangular form. "Newick" is the Newick representation of the tree.
"tip.label" is the character vector of tip labels. "tip.label.n" is the number of tips.
a matching representation of the phylogenetic tree corresponding to the input
Kem Phillips <[email protected]
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
P.W. Diaconis and S. Holmes, Matchings and Phylogenetic Trees, Proc. Natl. Acad. Sci., 1998, 95(25), 14600-14602
functions toMoment and toNewick
# create a Newick object exam.Newick <- "(((a,b),c),d);" # convert to a moment L-matrix exam.moment <- toMoment(exam.Newick) # convert to matching format exam.matching <- toMatching(exam.moment)# create a Newick object exam.Newick <- "(((a,b),c),d);" # convert to a moment L-matrix exam.moment <- toMoment(exam.Newick) # convert to matching format exam.matching <- toMatching(exam.moment)
Converts a tree from Newick or matching to moment format
toMoment(inputobject, tip.label = NULL)toMoment(inputobject, tip.label = NULL)
inputobject |
a tree in Newick format or a |
tip.label |
rearranged labels for tips; these must be the original labels |
The L-matrix class consists of $5$ components: "L" is the L-matrix in square form. "L.ut" is the L-matrix in upper-triangular form. "Newick" is the Newick representation of the tree. "tip.label" is the character vector of tip labels. "tip.label.n" is the number of tips.
a moment L-matrix corresponding to the input phylogentic tree object
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
J. Felsenstein, The Newick tree format, 1990, http://evolution.genetics.washington.edu/phylip/newicktree.html
P.W. Diaconis and S. Holmes, Matchings and Phylogenetic Trees, Proc. Natl. Acad. Sci., 1998, 95(25), 14600-14602
functions toNewick and toMatching
# create a Newick object exam.Newick <- "(((a,b),c),d);" # convert to a moment L-matrix exam.moment <- toMoment(exam.Newick) # convert to matching object exam.matching <- toMatching(exam.moment) # convert back to moment object backto.moment <- toMoment(exam.matching)# create a Newick object exam.Newick <- "(((a,b),c),d);" # convert to a moment L-matrix exam.moment <- toMoment(exam.Newick) # convert to matching object exam.matching <- toMatching(exam.moment) # convert back to moment object backto.moment <- toMoment(exam.matching)
function converts a tree in moment format to Newick format.
The input can be an L-matrix object,
a square $L$ matrix,
or an $L$ matrix in reduced upper-triangular (vector) form.
The toNewick function sets its list output to class L-Newick,
which has $5$ components,
including the tree in Newick format.
toNewick(L, type = NULL, tip.label = NULL)toNewick(L, type = NULL, tip.label = NULL)
L |
L can be an |
type |
if L is not a |
tip.label |
Character vector containing labels for tips. If null, labels default to "a"-"z" it at most 26; otherwise, 3-letter labels of the form "aaa", "aab",... |
An L-matrix object is a list with 5 components: "L" is the L-matrix in square form. "L.ut" is the L-matrix in upper-triangular form. "Newick" is the Newick representation of the tree.
"tip.label" is the character vector of tip labels. "tip.label.n" is the number of tips.
a Newick representation of the phylogenetic tree corresponding to the input
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
J. Felsenstein, The Newick tree format, 1990, http://evolution.genetics.washington.edu/phylip/newicktree.html
P.W. Diaconis and S. Holmes, Matchings and Phylogenetic Trees, Proc. Natl. Acad. Sci., 1998, 95(25), 14600-14602
functions toMoment and toMatching
# create a Newick object exam.Newick <- "(((a,b),c),d);" # convert to a moment L-matrix exam.moment <- toMoment(exam.Newick) # convert back to Newick format backto.Newick <- toNewick(exam.moment)# create a Newick object exam.Newick <- "(((a,b),c),d);" # convert to a moment L-matrix exam.moment <- toMoment(exam.Newick) # convert back to Newick format backto.Newick <- toNewick(exam.moment)
Produces an unsorted central moment object from a sorted object of class "moment".
Unsorted moments are those with exponents not in numeric order, e.g., m312.
tounsorted(moment,sorted.moment)tounsorted(moment,sorted.moment)
moment |
unsorted moment to obtain moment is in vector form, eg, c(3,1,2) |
sorted.moment |
sorted moment to use in obtaining unsorted moment |
The unsorted moment is obtained by resorting the rows and columns of the sorted moment sucessively.
A object of class 'moment', which is a list with three components:
moment |
the input moment vector |
representation |
a matrix containing the representation in terms of upper-triangular matrices |
coefficients |
the coefficients corresponding to the rows of the representation |
Kem Phillips <[email protected]>
K Phillips, Symbolic Computation of the Central Moments of the Multivariate Normal Distribution, Journal of Statistical Software, 2010.
multmoments
# obtain moment m312 from m123 tounsorted(c(3,1,2),callmultmoments(c(1,2,3)))# obtain moment m312 from m123 tounsorted(c(3,1,2),callmultmoments(c(1,2,3)))