UCSC - Bayesian Statistics: Techniques and Models
- Offered byCoursera
Bayesian Statistics: Techniques and Models at Coursera Overview
Duration | 30 hours |
Total fee | Free |
Mode of learning | Online |
Difficulty level | Intermediate |
Official Website | Explore Free Course  |
Credential | Certificate |
Bayesian Statistics: Techniques and Models at Coursera Highlights
- Shareable Certificate Earn a Certificate upon completion
- 100% online Start instantly and learn at your own schedule.
- Flexible deadlines Reset deadlines in accordance to your schedule.
- Intermediate Level
- Approx. 30 hours to complete
- English Subtitles: Arabic, French, Portuguese (European), Italian, Vietnamese, German, Russian, English, Spanish
Bayesian Statistics: Techniques and Models at Coursera Course details
- This is the second of a two-course sequence introducing the fundamentals of Bayesian statistics. It builds on the course Bayesian Statistics: From Concept to Data Analysis, which introduces Bayesian methods through use of simple conjugate models. Real-world data often require more sophisticated models to reach realistic conclusions. This course aims to expand our ?Bayesian toolbox? with more general models, and computational techniques to fit them. In particular, we will introduce Markov chain Monte Carlo (MCMC) methods, which allow sampling from posterior distributions that have no analytical solution. We will use the open-source, freely available software R (some experience is assumed, e.g., completing the previous course in R) and JAGS (no experience required). We will learn how to construct, fit, assess, and compare Bayesian statistical models to answer scientific questions involving continuous, binary, and count data. This course combines lecture videos, computer demonstrations, readings, exercises, and discussion boards to create an active learning experience. The lectures provide some of the basic mathematical development, explanations of the statistical modeling process, and a few basic modeling techniques commonly used by statisticians. Computer demonstrations provide concrete, practical walkthroughs. Completion of this course will give you access to a wide range of Bayesian analytical tools, customizable to your data.
Bayesian Statistics: Techniques and Models at Coursera Curriculum
Statistical modeling and Monte Carlo estimation
Course introduction
Objectives
Modeling process
Components of Bayesian models
Model specification
Posterior derivation
Non-conjugate models
Monte Carlo integration
Monte Carlo error and marginalization
Computing examples
Computing Monte Carlo error
Module 1 assignments and materials
Reference: Common probability distributions
Code for Lesson 3
Markov chains
Lesson 1
Lesson 2
Lesson 3
Markov chains
Markov chain Monte Carlo (MCMC)
Algorithm
Demonstration
Random walk example, Part 1
Random walk example, Part 2
Download, install, setup
Model writing, running, and post-processing
Multiple parameter sampling and full conditional distributions
Conditionally conjugate prior example with Normal likelihood
Computing example with Normal likelihood
Trace plots, autocorrelation
Multiple chains, burn-in, Gelman-Rubin diagnostic
Module 2 assignments and materials
Code for Lesson 4
Alternative MCMC software
Code from JAGS introduction
Code for Lesson 5
Autocorrelation
Code for Lesson 6
Lesson 4
Lesson 5
Lesson 6
MCMC
Common statistical models
Introduction to linear regression
Setup in R
JAGS model (linear regression)
Model checking
Alternative models
Deviance information criterion (DIC)
Introduction to ANOVA
One way model using JAGS
Introduction to logistic regression
JAGS model (logistic regression)
Prediction
Module 3 assignments and materials
Code for Lesson 7
Code for Lesson 8
Code for Lesson 9
Multiple factor ANOVA
Lesson 7 Part A
Lesson 7 Part B
Lesson 8
Lesson 9
Common models and multiple factor ANOVA
Count data and hierarchical modeling
Introduction to Poisson regression
JAGS model (Poisson regression)
Predictive distributions
Correlated data
Prior predictive simulation
JAGS model and model checking (hierarchical modeling)
Posterior predictive simulation
Linear regression example
Linear regression example in JAGS
Mixture model in JAGS
Module 4 assignments and materials
Prior sensitivity analysis
Code for Lesson 10
Normal hierarchical model
Applications of hierarchical modeling
Code and data for Lesson 11
Mixture model introduction, data, and code
Lesson 10
Lesson 11 Part A
Lesson 11 Part B
Predictive distributions and mixture models
Capstone project
Course conclusion
Further reading and acknowledgements
Other courses offered by Coursera
Student Forum
Anything you would want to ask experts?