SteadyState

A small project that analyzes the convergence of several Markov Chains.

Overview

• We first examine two Markov Chains with significantly different transition matrices in convergence.py.
• Then, we extend the analysis to three additional chains with distinct behaviors in extra_transitions.py.
• We compare how different initial distributions affect convergence in extra_initial_dists.py.
• Finally, we simulate Markov Chains step by step to observe empirical behavior in simulation.py.

This project is part of my Bayesian Networks and Hidden Markov Models class.

Built together with my colleague Mara Fodor.

---

Setup

Poetry (Recommended)

The project uses Poetry for dependency management.

After installing Poetry:

1. Install project dependencies:

   poetry install

2. Activate the project environment:

   poetry shell

3. Run the different scripts:

   python -m steadystate.convergence
   python -m steadystate.extra_transitions
   python -m steadystate.extra_initial_dists
   python -m steadystate.simulation