Mixture(w, comp_dists, *args, **kwargs) Mixture log-likelihood
NormalMixture(w, mu[, comp_shape]) Normal mixture log-likelihood
class pymc3.distributions.mixture.Mixture(w, comp_dists, *args, **kwargs)

Mixture log-likelihood

Often used to model subpopulation heterogeneity

\[f(x \mid w, \theta) = \sum_{i = 1}^n w_i f_i(x \mid \theta_i)\]
Support \(\cap_{i = 1}^n \textrm{support}(f_i)\)
Mean \(\sum_{i = 1}^n w_i \mu_i\)
  • w (array of floats) – w >= 0 and w <= 1 the mixture weights
  • comp_dists (multidimensional PyMC3 distribution (e.g. pm.Poisson.dist(…))) – or iterable of one-dimensional PyMC3 distributions the component distributions \(f_1, \ldots, f_n\)


# 2-Mixture Poisson distribution
with pm.Model() as model:
    lam = pm.Exponential('lam', lam=1, shape=(2,))  # `shape=(2,)` indicates two mixtures.

    # As we just need the logp, rather than add a RV to the model, we need to call .dist()
    components = pm.Poisson.dist(mu=lam, shape=(2,))

    w = pm.Dirichlet('w', a=np.array([1, 1]))  # two mixture component weights.

    like = pm.Mixture('like', w=w, comp_dists=components, observed=data)

# 2-Mixture Poisson using iterable of distributions.
with pm.Model() as model:
    lam1 = pm.Exponential('lam1', lam=1)
    lam2 = pm.Exponential('lam2', lam=1)

    pois1 = pm.Poisson.dist(mu=lam1)
    pois2 = pm.Poisson.dist(mu=lam2)

    w = pm.Dirichlet('w', a=np.array([1, 1]))

    like = pm.Mixture('like', w=w, comp_dists = [pois1, pois2], observed=data)
class pymc3.distributions.mixture.NormalMixture(w, mu, comp_shape=(), *args, **kwargs)

Normal mixture log-likelihood

\[f(x \mid w, \mu, \sigma^2) = \sum_{i = 1}^n w_i N(x \mid \mu_i, \sigma^2_i)\]
Support \(x \in \mathbb{R}\)
Mean \(\sum_{i = 1}^n w_i \mu_i\)
Variance \(\sum_{i = 1}^n w_i^2 \sigma^2_i\)
  • w (array of floats) – w >= 0 and w <= 1 the mixture weights
  • mu (array of floats) – the component means
  • sd (array of floats) – the component standard deviations
  • tau (array of floats) – the component precisions
  • comp_shape (shape of the Normal component) – notice that it should be different than the shape of the mixture distribution, with one axis being the number of components.
  • Note (You only have to pass in sd or tau, but not both.) –

Determine if all distributions in comp_dists are discrete