Statistics

This subpackage provides probability distributions, sampling algorithms, and statistical utilities used throughout PyOED.

Distributions model error sources (prior, observation noise) and serve as building blocks for Bayesian inversion and stochastic OED. Samplers (including MCMC) generate samples from these distributions, while proposals drive the Metropolis-Hastings accept/reject step.

Key Classes at a Glance

pyoed.stats.Proposal	Base class for Proposal (algorithms to propose samples which can be accepted or rejected by the MH step in MCMC).
pyoed.stats.Sampler	Base class for Samplers (algorithms to generate samples from a predefined distribution).

Class Hierarchy

Statistics abstract base class hierarchy

• Stats Package Core
  • Distributions
  • Sampling
• Probability Distributions
  • Multivariate Bernoulli Model
  • Poisson Binomial Distribution
  • Conditional Bernoulli Model
  • Trajectory Distribution (Markovian)
  • Combinatorial Functions
• Sampling
  • Rejection Sampling
  • Markov Chain Monte Carlo Sampling
  • Proposals
• Statistical utilities and helper functions