statsmodels.distributions.bernstein.BernsteinDistribution

class statsmodels.distributions.bernstein.BernsteinDistribution(cdf_grid)

Distribution based on Bernstein Polynomials on unit hypercube.

Parameters:

cdf_gridarray_like

cdf values on a equal spaced grid of the unit hypercube [0, 1]^d. The dimension of the arrays define how many random variables are included in the multivariate distribution.

Attributes:

cdf_gridndarray

Grid of cdf values.

prob_gridndarray

Grid of cell or bin probabilities.

k_dimint

Number of components, dimension of the random variable.

k_gridtuple of int

Shape of cdf_grid.

k_grid_productint

Total number of bins in the grid.

_grid_Grid

Instance with grid helper methods and attributes.

Methods

cdf(x)	cdf values evaluated at x.
from_data(data, k_bins)	Create distribution instance from data using histogram binning.
get_marginal(idx)	Get marginal BernsteinDistribution.
pdf(x)	pdf values evaluated at x.
rvs(nobs[, rng])	Generate random numbers from distribution.

Methods

cdf(x)	cdf values evaluated at x.
from_data(data, k_bins)	Create distribution instance from data using histogram binning.
get_marginal(idx)	Get marginal BernsteinDistribution.
pdf(x)	pdf values evaluated at x.
rvs(nobs[, rng])	Generate random numbers from distribution.

Properties

prob_grid