rspd_matrix

Generate a random symetric positive definite matrix. The generator just
draw matrix of the form O * diag(positive values) * t(O), where O is an
orthogonal matrix from ROOPSD::rorthogonal_group. Note that the parameter
gen = stats::rexp draw positive eigen values, but the code do not control
if eigen values are positive. So you can accept negative eigen values using
another generators.

Statistical distribution in OOP (Object Oriented Programming) way.
This package proposes a R6 class interface to classic statistical
distribution, and new distributions can be easily added with the
class AbstractDist. A useful point is the generic fit() method for
each class, which uses a maximum likelihood estimation to find the
parameters of a dataset, see, e.g. Hastie, T. and al (2009)
<isbn:978-0-387-84857-0>. Furthermore, the rv_histogram class
gives a non-parametric fit, with the same accessors that for the
classic distribution. Finally, three random generators useful to
build synthetic data are given: a multivariate normal generator, an
orthogonal matrix generator, and a symmetric positive definite
matrix generator, see Mezzadri, F. (2007)
<arXiv:math-ph/0609050>.

Yoann Robin

ROOPSD

R Object Oriented Programming for Statistical Distribution

rspd_matrix function

<dl><dt>d</dt>
<dd>[integer] Dimension of the matrix</dd>
<dt>n</dt>
<dd>[integer] numbers of samples drawn</dd>
<dt>sort_eigenvalues</dt>
<dd>[bool] If eigen values (i.e. variance) are sorted</dd>
<dt>gen</dt>
<dd>[function] Eigenvalues generator</dd></dl>

Arguments

rspd_matrix — rspd_matrix

<dl>

<dt>d</dt>
<dd>[integer] Dimension of the matrix</dd>


<dt>n</dt>
<dd>[integer] numbers of samples drawn</dd>


<dt>sort_eigenvalues</dt>
<dd>[bool] If eigen values (i.e. variance) are sorted</dd>


<dt>gen</dt>
<dd>[function] Eigenvalues generator</dd>

</dl>

rspd_matrix: rspd_matrix

Description

Usage

Value

Arguments

Examples