Printable Handouts

PDF

Navigable Slide Index

1. Introduction
2. Review of Markov chains
3. Transition matrix
4. Stationary distribution (1)
5. Stationary distribution (2)
6. Time reversible chain (1)
7. Time reversible chain (2)
8. Stationary and time reversibility
9. Example
10. Principle number 1
11. Metropolis methods
12. Discrete case: Metropolis-Hastings algorithm (1)
13. Discrete case: Metropolis-Hastings algorithm (2)
14. Metropolis-Hastings algorithm - why do we repeat
15. Time reversible with respect to pie
16. Metropolis-Hastings algorithm example
17. Continuous Metropolis-Hastings
18. Principle number 2
19. Independence chain
20. Random walk (rw) chains
21. Independence vs. rw chains
22. Choosing a step size for the rw chain (1)
23. Relative numerical efficiency (1)
24. Relative numerical efficiency (2)
25. Choosing a step size for the rw chain (2)
26. The Gibbs sampler (1)
27. The Gibbs sampler (2)
28. Principle number 3
29. Logit model
30. Logit model-Hessian
31. Logit model MCMC algorithms
32. Scaling rw Metropolis
33. Comparison of indep / rw Metropolis
34. rmnlIndepMetrop (1)
35. rmnlIndepMetrop (2)
36. Time series of all the draws of beta
37. Quantile regression
38. Asymmetric Laplace distribution
39. Bayesian quantile regression
40. Quantile R code
41. Summary

Topics Covered

• Review of Markov chains
• Stationary distribution
• Time reversible chains
• Metropolis methods
• Discrete case: Metropolis-Hastings algorithm
• Independence chains
• Random walk chains
• Relative numerical efficiency
• The Gibbs sampler
• Logit model
• Scaling rw metropolis
• Quantile regression
• Asymmetric Laplace distribution

Links

Series:

• Bayesian Analysis in Marketing

Categories:

• Finance, Accounting & Economics
• Marketing & Sales

Talk Citation

Publication History

• Published on January 27, 2010