Skip to content

ptjahjadi/MCMC-Methods

Repository files navigation

MCMC-Methods

Aim

This assignment aims to develop a Gibbs sampling and Metropolis Hastings Algorithm to sample from a specified probability distribution. This is the final project of the Modern Applied Statistics (MAST30027) subject at the University of Melbourne.

Summary

Markov Chain Monte Carlo (MCMC) methods are powerful statistical algorithms to sample from a probability distribution. In this project, both Gibbs sampling and Metropolis Hastings Algorithm are implemented to sample using MCMC methods. In order to see the entire distribution, including the first few samples that have not converged to the proposed distribution, no burn-in is used in the code. However, burn-in may be implemented as necessary.

Guide

The folder "Visualisations" contains all necessary visualisations for this project.

The problem is specified in "2019_Assignment4_v2.pdf".

The report can be found in "MAST30027 A4.pdf".

Every code used during this project is located in "Rcode.R".

Built With

  • R
  • LaTeX

Special Thanks

  • Dr. Heejung Shim and Qiuyi Li
  • The University of Melbourne

About

This assignment aims to develop a Gibbs sampling and Metropolis Hastings Algorithm to sample from a specified probability distribution.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors