https://github.com/ondrejbiza/mcmc

Markov Chain Monte Carlo methods.
https://github.com/ondrejbiza/mcmc

mcmc mcmc-sampler metropolis-hastings probability sampling statistics

Last synced: about 1 month ago
JSON representation

Markov Chain Monte Carlo methods.

• Host: GitHub
• URL: https://github.com/ondrejbiza/mcmc
• Owner: ondrejbiza
• License: mit
• Created: 2019-03-24T14:03:39.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2019-03-24T15:15:07.000Z (almost 6 years ago)
• Last Synced: 2024-11-10T23:12:32.586Z (3 months ago)
• Topics: mcmc, mcmc-sampler, metropolis-hastings, probability, sampling, statistics
• Language: Python
• Size: 317 KB
• Stars: 1
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Markov Chain Monte Carlo methods

This repository replicates the figures from [C. Andrieu's et al. tutorial on MCMC methods](https://link.springer.com/article/10.1023/A:1020281327116).

## The Metropolis-Hastings algorithm

Changing iterations:

![100_iter](results/mh_100_it_100_var.png) 

![500_iter](results/mh_500_it_100_var.png)

![1000_iter](results/mh_1000_it_100_var.png)

![5000_iter](results/mh_5000_it_100_var.png)

Changing variance of the proposal distribution:

![0.1_var](results/mh_5000_it_0.1_var.png) 

![1_var](results/mh_5000_it_1_var.png)

![100_var](results/mh_5000_it_100_var.png)

![1000_var](results/mh_5000_it_1000_var.png)

![2000_var](results/mh_5000_it_2000_var.png)