StochX

A Python library for stochastic processes and applied probability.

The library currently covers:

• discrete-time Markov chains (DTMC);
• continuous-time Markov chains (CTMC);
• Poisson processes;
• birth-death processes;
• finite probability spaces, partitions, random variables, and conditional expectation;
• filtrations, martingales, stopping times, and stopped processes.

Installation

pip install stochx

Quick example

Create a discrete-time Markov chain from its transition matrix:

import numpy as np
from stochx.stochastic import MarkovChain

P = np.array([
    [0.8, 0.2],
    [0.3, 0.7],
])

chain = MarkovChain(P, states=["A", "B"])

print(chain.states)
print(chain.transition_matrix)
print(chain.stationary_distribution())

Documentation

The documentation is divided into three parts.

Course material

The course chapters present the mathematical definitions, notation, results, and hypotheses used throughout the stochastic-process material. The mathematics is kept separate from the software reference.

Package / API

The API Reference documents the public StochX classes, functions, properties, methods, validation rules, and numerical behavior.

Examples

Worked examples show how the mathematical objects are represented and used in Python.

Supported stochastic models

StochX — stochastic-process mathematics implemented as a Python library.