Random matrix package for Julia.
This extends the Distributions package to provide methods for working with matrix-valued random variables, a.k.a. random matrices. State of the art methods for computing random matrix samples and their associated distributions are provided.
The names of the various ensembles can vary widely across disciplines. Where possible, synonyms are listed.
Additional functionality is provided when these optional packages are installed:
- Symbolic manipulation of Haar matrices with GSL.jl
Much of classical random matrix theory has focused on matrices with matrix elements comprised of
independently and identically distributed (iid) real, complex or quaternionic Gaussians.
(Traditionally, these are associated with a parameter beta
tracking the number of independent
real random variables per matrix element, i.e. beta=1,2,4
respectively. This is also referred
to as the Dyson 3-fold way.)
Methods are provided for calculating random variates (samples) and various properties of these
random matrices.
The hierarchy of dense matrices provided are
- Ginibre ensemble - all matrix elements are iid with no global symmetry
- Hermite ensemble - one global symmetry
- Gaussian orthogonal ensemble (GOE,
beta=1
) - real and symmetric - Gaussian unitary ensemble (GUE,
beta=2
) - complex and Hermitian - Gaussian symplectic ensemble (GSE,
beta=4
) - quaternionic and self-dual
- Gaussian orthogonal ensemble (GOE,
- Circular ensemble - uniformly distributed with
|det|=1
- Circular orthogonal ensemble (COE,
beta=1
) - Circular unitary ensemble (CUE,
beta=2
) - Circular symplectic ensemble (CSE,
beta=4
)
- Circular orthogonal ensemble (COE,
- Laguerre matrices = white Wishart matrices
- Jacobi matrices = MANOVA matrices
Unless otherwise specified, beta=1,2,4
are supported. For the symplectic matrices beta=4
,
the 2x2 outer block-diagonal complex representation USp(2N)
is used.
Given eigenvalues lambda
and the beta
parameter of the random matrix distribution:
VandermondeDeterminant(lambda, beta)
computes the Vandermonde determinantHermiteJPDF(lambda, beta)
computes the jpdf for the Hermite ensembleLaguerreJPDF(lambda, n, beta)
computes the jpdf for the Laguerre(n) ensembleJacobiJPDF(lambda, n1, n2, beta)
computes the jpdf for the Jacobi(n1, n2) ensemble
Constructs samples of random matrices corresponding to the classical Gaussian Hermite, Laguerre(m) and Jacobi(m1, m2) ensembles.
-
GaussianHermiteMatrix(n, beta)
,GaussianLaguerreMatrix(n, m, beta)
,GaussianJacobiMatrix(n, m1, m2, beta)
each construct a sample densen
xn
matrix for the corresponding matrix ensemble withbeta=1,2,4
-
GaussianHermiteTridiagonalMatrix(n, beta)
,GaussianLaguerreTridiagonalMatrix(n, m, beta)
,GaussianJacobiSparseMatrix(n, m1, m2, beta)
each construct a sparsen
xn
matrix for the corresponding matrix ensemble for arbitrary positive finitebeta
.GaussianHermiteTridiagonalMatrix(n, Inf)
is also allowed. These sampled matrices have the same eigenvalues as above but are much faster to diagonalize oweing to their sparsity. They also extend Dyson's threefold way to arbitrarybeta
. -
GaussianHermiteSamples(n, beta)
,GaussianLaguerreSamples(n, m, beta)
,GaussianJacobiSamples(n, m1, m2, beta)
return a set ofn
eigenvalues from the sparse random matrix samples -
HaarMatrix(n, beta)
Generates a random matrix from thebeta
-circular ensemble. -
HaarMatrix(n, beta, correction)
provides fine-grained control of what kind of correction is applied to the raw QR decomposition. By default,correction=1
(Edelman's correction) is used. Other valid values are0
(no correction) and2
(Mezzadri's correction). -
NeedsPiecewiseCorrection()
implements a simple test to see if a correction is necessary.
The parameters m
, m1
, m2
refer to the number to independent "data" degrees of freedom.
For the dense samples these must be Integer
s but can be Real
s for the rest.
Allows for manipulations of formal power series (fps) and formal Laurent series (fLs), which come in handy for the computation of free cumulants.
FormalPowerSeries
: power series with coefficients allowed only for non-negative integer powersFormalLaurentSeries
: power series with coefficients allowed for all integer powers
- basic arithmetic operations
==
,+
,-
,^
*
computes the Cauchy product (discrete convolution).*
computes the Hadamard product (elementwise multiplication)compose(P,Q)
computes the series composition P.Qderivative
computes the series derivativereciprocal
computes the series reciprocal
trim(P)
removes extraneous zeroes in the internal representation ofP
isalmostunit(P)
determines ifP
is an almost unit seriesisconstant(P)
determines ifP
is a constant seriesisnonunit(P)
determines ifP
is a non-unit seriesisunit(P)
determines ifP
is a unit seriesMatrixForm(P)
returns a matrix representation ofP
as an upper triangular Toeplitz matrixtovector
returns the series coefficients
Famous distributions in random matrix theory
Semicircle
provides the semicircle distributionTracyWidom
computes the Tracy-Widom density distribution by brute-force integration of the Painlevé II equation
hist_eig
computes the histogram of eigenvalues of a matrix using the method of Sturm sequences. This is recommended forSymTridiagonal
matrices as it is significantly faster thanhist(eigvals())
This is also implemented for dense matrices, but it is pretty slow and not really practical.
Julia iterators for stochastic operators.
All subtypes of StochasticProcess
contain at least one field, dt
,
representing the time interval being discretized over.
The available StochasticProcess
es are
BrownianProcess(dt)
: Brownian random walk. The state of the iterator is the cumulative displacement of the random walk.WhiteNoiseProcess(dt)
: White noise. The value of this iterator israndn()*dt
. The state associated with this iterator isnothing
.StochasticAiryProcess(dt, beta)
: stochastic Airy process with real positivebeta
. The value of this iterator in the limit of an infinite number of iterations is known to follow thebeta
-Tracy-Widom law. The state associated with this iteratior is aSymTridiagonal
matrix whose largest eigenvalue is the value of this process.
InvariantEnsemble(str,n)
supports n x n unitary invariant ensemble
with distribution. This has been moved to separate package InvariantEnsembles.jl
-
James Albrecht, Cy Chan, and Alan Edelman, "Sturm Sequences and Random Eigenvalue Distributions", Foundations of Computational Mathematics, vol. 9 iss. 4 (2009), pp 461-483. [pdf] [doi]
-
Ioana Dumitriu and Alan Edelman, "Matrix Models for Beta Ensembles", Journal of Mathematical Physics, vol. 43 no. 11 (2002), pp. 5830-5547 [[doi]](http://dx.doi.org/doi: 10.1063/1.1507823) arXiv:math-ph/0206043
-
Alan Edelman, Per-Olof Persson and Brian D Sutton, "The fourfold way", Journal of Mathematical Physics, submitted (2013). [pdf]
-
Alan Edelman and Brian D. Sutton, "The beta-Jacobi matrix model, the CS decomposition, and generalized singular value problems", Foundations of Computational Mathematics, vol. 8 iss. 2 (2008), pp 259-285. [pdf] [doi]
-
Peter Henrici, Applied and Computational Complex Analysis, Volume I: Power Series---Integration---Conformal Mapping---Location of Zeros, Wiley-Interscience: New York, 1974 [worldcat]
-
Frank Mezzadri, "How to generate random matrices from the classical compact groups", Notices of the AMS, vol. 54 (2007), pp592-604 [arXiv]