statsmodels.distributions.mixture_rvs.mv_mixture_rvs#

statsmodels.distributions.mixture_rvs.mv_mixture_rvs(prob, size, dist, nvars, rng=None, **kwargs)[source]#

Sample from a mixture of multivariate distributions.

Parameters:
probarray_like

Probability of sampling from each distribution in dist

sizeint

The length of the returned sample.

distarray_like

An iterable of distributions instances with callable method rvs.

nvarsint

dimension of the multivariate distribution, could be inferred instead

rng{None, numpy.random.Generator, numpy.random.RandomState}, optional

If rng is None, the global (legacy) NumPy random state is used. If rng is already a Generator or RandomState instance, that instance is used.

kwargstuple of dicts, optional

ignored

Examples

Say we want 2000 random variables from mixture of normals with two multivariate normal distributions, and we want to sample from the first with probability .4 and the second with probability .6.

import statsmodels.sandbox.distributions.mv_normal as mvd

cov3 = np.array([[ 1. , 0.5 , 0.75],

[ 0.5 , 1.5 , 0.6 ], [ 0.75, 0.6 , 2. ]])

mu = np.array([-1, 0.0, 2.0]) mu2 = np.array([4, 2.0, 2.0]) mvn3 = mvd.MVNormal(mu, cov3) mvn32 = mvd.MVNormal(mu2, cov3/2., 4) rvs = mix.mv_mixture_rvs([0.4, 0.6], 2000, [mvn3, mvn32], 3)