MvNormal(mu[, cov, tau, gpu_compat]) Multivariate normal log-likelihood.
Wishart(n, V, \*args, \*\*kwargs) Wishart log-likelihood.
LKJCholeskyCov(n, eta, sd_dist, \*args, \*\*kwargs) Covariance matrix with LKJ distributed correlations.
LKJCorr(n, p[, transform]) The LKJ (Lewandowski, Kurowicka and Joe) log-likelihood.
Multinomial(n, p, \*args, \*\*kwargs) Multinomial log-likelihood.
Dirichlet(a[, transform]) Dirichlet log-likelihood.
class pymc3.distributions.multivariate.MvNormal(mu, cov=None, tau=None, gpu_compat=False, *args, **kwargs)

Multivariate normal log-likelihood.

\[f(x \mid \pi, T) = \frac{|T|^{1/2}}{(2\pi)^{1/2}} \exp\left\{ -\frac{1}{2} (x-\mu)^{\prime} T (x-\mu) \right\}\]
Support \(x \in \mathbb{R}^k\)
Mean \(\mu\)
Variance \(T^{-1}\)
  • mu (array) – Vector of means.
  • cov (array) – Covariance matrix. Not required if tau is passed.
  • tau (array) – Precision matrix. Not required if tau is passed.
  • Flags
  • ----------
  • gpu_compat (False, because LogDet is not GPU compatible yet.) – If this is set as true, the GPU compatible (but numerically unstable) log(det) is used.
class pymc3.distributions.multivariate.MvStudentT(nu, Sigma, mu=None, *args, **kwargs)

Multivariate Student-T log-likelihood.

\[f(\mathbf{x}| \nu,\mu,\Sigma) = \frac{\Gamma\left[(\nu+p)/2\right]}{\Gamma(\nu/2)\nu^{p/2}\pi^{p/2}\left|{\Sigma}\right|^{1/2}\left[1+\frac{1}{\nu}({\mathbf x}-{\mu})^T{\Sigma}^{-1}({\mathbf x}-{\mu})\right]^{(\nu+p)/2}}\]
Support \(x \in \mathbb{R}^k\)
Mean \(\mu\) if \(\nu > 1\) else undefined
if \(\nu>2\) else undefined
  • nu (int) – Degrees of freedom.
  • Sigma (matrix) – Covariance matrix.
  • mu (array) – Vector of means.
class pymc3.distributions.multivariate.Dirichlet(a, transform=<pymc3.distributions.transforms.StickBreaking object>, *args, **kwargs)

Dirichlet log-likelihood.

\[f(\mathbf{x}) = \frac{\Gamma(\sum_{i=1}^k \theta_i)}{\prod \Gamma(\theta_i)} \prod_{i=1}^{k-1} x_i^{\theta_i - 1} \left(1-\sum_{i=1}^{k-1}x_i\right)^\theta_k\]
Support \(x_i \in (0, 1)\) for \(i \in \{1, \ldots, K\}\) such that \(\sum x_i = 1\)
Mean \(\dfrac{a_i}{\sum a_i}\)
Variance \(\dfrac{a_i - \sum a_0}{a_0^2 (a_0 + 1)}\) where \(a_0 = \sum a_i\)
Parameters:a (array) – Concentration parameters (a > 0).


Only the first k-1 elements of x are expected. Can be used as a parent of Multinomial and Categorical nevertheless.

class pymc3.distributions.multivariate.Multinomial(n, p, *args, **kwargs)

Multinomial log-likelihood.

Generalizes binomial distribution, but instead of each trial resulting in “success” or “failure”, each one results in exactly one of some fixed finite number k of possible outcomes over n independent trials. ‘x[i]’ indicates the number of times outcome number i was observed over the n trials.

\[f(x \mid n, p) = \frac{n!}{\prod_{i=1}^k x_i!} \prod_{i=1}^k p_i^{x_i}\]
Support \(x \in \{0, 1, \ldots, n\}\) such that \(\sum x_i = n\)
Mean \(n p_i\)
Variance \(n p_i (1 - p_i)\)
Covariance \(-n p_i p_j\) for \(i \ne j\)
  • n (int or array) – Number of trials (n > 0).
  • p (one- or two-dimensional array) – Probability of each one of the different outcomes. Elements must be non-negative and sum to 1 along the last axis. They will be automatically rescaled otherwise.
class pymc3.distributions.multivariate.Wishart(n, V, *args, **kwargs)

Wishart log-likelihood.

The Wishart distribution is the probability distribution of the maximum-likelihood estimator (MLE) of the precision matrix of a multivariate normal distribution. If V=1, the distribution is identical to the chi-square distribution with n degrees of freedom.

\[f(X \mid n, T) = \frac{{\mid T \mid}^{n/2}{\mid X \mid}^{(n-k-1)/2}}{2^{nk/2} \Gamma_p(n/2)} \exp\left\{ -\frac{1}{2} Tr(TX) \right\}\]

where \(k\) is the rank of \(X\).

Support \(X(p x p)\) positive definite matrix
Mean \(n V\)
Variance \(n (v_{ij}^2 + v_{ii} v_{jj})\)
  • n (int) – Degrees of freedom, > 0.
  • V (array) – p x p positive definite matrix.


This distribution is unusable in a PyMC3 model. You should instead use WishartBartlett or LKJCorr.

pymc3.distributions.multivariate.WishartBartlett(name, S, nu, is_cholesky=False, return_cholesky=False, testval=None)

Bartlett decomposition of the Wishart distribution. As the Wishart distribution requires the matrix to be symmetric positive semi-definite it is impossible for MCMC to ever propose acceptable matrices.

Instead, we can use the Barlett decomposition which samples a lower diagonal matrix. Specifically:

\[ \begin{align}\begin{aligned}\begin{split}\text{If} L \sim \begin{pmatrix} \sqrt{c_1} & 0 & 0 \\ z_{21} & \sqrt{c_2} & 0 \\ z_{31} & z_{32} & \sqrt{c_3} \end{pmatrix}\end{split}\\\begin{split}\text{with} c_i \sim \chi^2(n-i+1) \text{ and } n_{ij} \sim \mathcal{N}(0, 1), \text{then} \\ L \times A \times A.T \times L.T \sim \text{Wishart}(L \times L.T, \nu)\end{split}\end{aligned}\end{align} \]

See for more information.

  • S (ndarray) – p x p positive definite matrix Or: p x p lower-triangular matrix that is the Cholesky factor of the covariance matrix.
  • nu (int) – Degrees of freedom, > dim(S).
  • is_cholesky (bool (default=False)) – Input matrix S is already Cholesky decomposed as S.T * S
  • return_cholesky (bool (default=False)) – Only return the Cholesky decomposed matrix.
  • testval (ndarray) – p x p positive definite matrix used to initialize


This is not a standard Distribution class but follows a similar interface. Besides the Wishart distribution, it will add RVs c and z to your model which make up the matrix.

class pymc3.distributions.multivariate.LKJCorr(n, p, transform='interval', *args, **kwargs)

The LKJ (Lewandowski, Kurowicka and Joe) log-likelihood.

The LKJ distribution is a prior distribution for correlation matrices. If n = 1 this corresponds to the uniform distribution over correlation matrices. For n -> oo the LKJ prior approaches the identity matrix.

Support Upper triangular matrix with values in [-1, 1]
  • n (float) – Shape parameter (n > 0). Uniform distribution at n = 1.
  • p (int) – Dimension of correlation matrix (p > 0).


This implementation only returns the values of the upper triangular matrix excluding the diagonal. Here is a schematic for p = 5, showing the indexes of the elements:

[[- 0 1 2 3]
 [- - 4 5 6]
 [- - - 7 8]
 [- - - - 9]
 [- - - - -]]


[LKJ2009]Lewandowski, D., Kurowicka, D. and Joe, H. (2009). “Generating random correlation matrices based on vines and extended onion method.” Journal of multivariate analysis, 100(9), pp.1989-2001.
class pymc3.distributions.multivariate.LKJCholeskyCov(n, eta, sd_dist, *args, **kwargs)

Covariance matrix with LKJ distributed correlations.

This defines a distribution over cholesky decomposed covariance matrices, such that the underlying correlation matrices follow an LKJ distribution [1] and the standard deviations follow an arbitray distribution specified by the user.

  • n (int) – The number of rows of the covariance matrix.
  • eta (float) – The shape parameter of the LKJ distribution. A value of one implies a uniform distribution of the correlation matrices; larger values put more weight on matrices with few correlations.
  • sd_dist (pm.Distribution) – A distribution for the standard deviations.


Since the cholesky factor is a lower triangular matrix, we use packed storge for the matrix: We store and return the values of the lower triangular matrix in a one-dimensional array, numbered by row:

[[0 - - -]
 [1 2 - -]
 [3 4 5 -]
 [6 7 8 9]]

You can use pm.expand_packed_triangular(packed_cov, lower=True) to convert this to a regular two-dimensional array.


with pm.Model() as model:
    # Note that we access the distribution for the standard
    # deviations, and do not create a new random variable.
    sd_dist = pm.HalfCauchy.dist(beta=2.5)
    packed_chol = pm.LKJCholeskyCov('chol_cov', 10, 2, sd_dist)

    # Define a new MvNormal with the given covariance
    vals = pm.MvNormal('vals', mu=np.zeros(10), packed_chol=packed_chol)

    # Or transform an uncorrelated normal:
    vals_raw = pm.Normal('vals_raw', mu=np.zeros(10), sd=1)
    chol = pm.expand_packed_triangular(10, packed_chol, lower=True)
    vals =, vals_raw)

    # Or compute the covariance matrix
    chol = pm.expand_packed_triangular(10, packed_chol, lower=True)
    cov =, chol.T)

    # Extract the standard deviations
    stds = tt.sqrt(tt.diag(cov))

In the unconstrained space all values of the cholesky factor are stored untransformed, except for the diagonal entries, where we use a log-transform to restrict them to positive values.

To correctly compute log-likelihoods for the standard deviations and the correlation matrix seperatly, we need to consider a second transformation: Given a cholesky factorization \(LL^T = \Sigma\) of a covariance matrix we can recover the standard deviations \(\sigma\) as the euclidean lengths of the rows of \(L\), and the cholesky factor of the correlation matrix as \(U = \text{diag}(\sigma)^{-1}L\). Since each row of \(U\) has length 1, we do not need to store the diagonal. We define a transformation \(\phi\) such that \(\phi(L)\) is the lower triangular matrix containing the standard deviations \(\sigma\) on the diagonal and the correlation matrix \(U\) below. In this form we can easily compute the different likelihoods seperatly, as the likelihood of the correlation matrix only depends on the values below the diagonal, and the likelihood of the standard deviation depends only on the diagonal values.

We still need the determinant of the jacobian of \(\phi^{-1}\). If we think of \(\phi\) as an automorphism on \(\mathbb{R}^{\tfrac{n(n+1)}{2}}\), where we order the dimensions as described in the notes above, the jacobian is a block-diagonal matrix, where each block corresponds to one row of \(U\). Each block has arrowhead shape, and we can compute the determinant of that as described in [2]. Since the determinant of a block-diagonal matrix is the product of the determinants of the blocks, we get

\[\text{det}(J_{\phi^{-1}}(U)) = \left[ \prod_{i=2}^N u_{ii}^{i - 1} L_{ii} \right]^{-1}\]


[1]Lewandowski, D., Kurowicka, D. and Joe, H. (2009). “Generating random correlation matrices based on vines and extended onion method.” Journal of multivariate analysis, 100(9), pp.1989-2001.
[2]J. M. isn’t a mathematician ( j-m-isnt-a-mathematician), Different approaches to evaluate this determinant, URL (version: 2012-04-14):