Probabilistic Programming and Monte Carlo Methods.

Again, the book Probabilistic Programming and Bayesian Methods for Hackers is a great free resource to start with if you want to get your hands dirty quickly.

• Markov Chain Monte Carlo & Gibbs Sampling

  Lecture notes. A mathy but readable description of MCMC methods and Gibb’s sampling as a way of sampling the posterior distribution in for use in Bayesian approaches.

• Marketing Analytics through Markov Chain

  Simple example of building a Markov Chain using a transition matrix supposed to represent customer states… e.g. did they make a purchase, were they aware of marketing campaign… etc.

• Understanding Monte Carlo Simulation

  Examples (with Python code) showing how to use Monte Carlo methods to sample from a distribution and also integrate a function, but for 1D numerical integrations, you may be better off using quadrature rules in my opinion.

• A Zero-Math Introduction to Markov Chain Monte Carlo Methods

  Does what it says on the tin. Good for building intuition about MCMC does.

• Markov Chain Monte Carlo Without all the Bullshit

  A basic description of MCMC methods using minimal jargon and some math but not lots of math.

Videos & Presentations

• An introduction to Gibbs sampling

  This video is part of a lecture course which closely follows the material covered in the book, “A Student’s Guide to Bayesian Statistics.” Uses a bivariate discrete probability distribution example to illustrate how Gibbs sampling works in practice.

• Introduction to Gibbs Sampling

  Slides from Duke’s STA 360 course: Bayesian Methods and Modern Statistics I’m not a big fan of slides without hearing the presentation, but they look readable. Covers Gibb’s sampling with the following distributions as examples: exponential, normal, and pareto.

• Markov Chain Monte Carlo and the Metropolis Algorithm

  Youtube: An introduction to the intuition of MCMC and implementation of the Metropolis algorithm. (35 minutes)

• MCMC: Markov Chain Monte Carlo

  A youtube lecture describing MCMC (20 minutes)

---