Computational Statistics EBook
An online Jupyter Book on Bayesian Data Analysis and Parameter Estimation Methods.
I created an online book on Computational Statistics using Jupyter Book. It’s designed to explain the mathematical principles behind key Bayesian Data Analysis and Parameter Estimation methods, complete with code implementations to illustrate the concepts in action. The EBook was used to explain topics and concepts to other students in Probability/Stastics/ Probabilistic Machine Learning Classes.
The primary goal was to create a practical, hands-on resource for anyone interested in the computational aspects of modern statistics.
Key Topics Covered
The book delves into a variety of essential topics, providing both theoretical background and functional code. Key areas include:
- Random Variable Generation: Techniques for sampling from various distributions.
- Monte Carlo Methods: Including Monte Carlo Integration for numerical approximation.
- Markov Chain Monte Carlo (MCMC): Detailed explanations of Metropolis-Hastings, Hamiltonian Monte Carlo (HMC), and Gibbs Sampling.
- Gaussian Processes: A non-parametric approach to regression and classification.
- Information Theory: Core concepts like entropy and Kullback-Leibler divergence.
- Sequential Monte Carlo (SMC): Also known as particle filters.
- Graphical Models: Representing probabilistic relationships between random variables.
Project Showcase
Here is a glimpse of some of the visualizations and code snippets included in the book:
