Printable Handouts

PDF

Navigable Slide Index

1. Introduction
2. Agenda
3. Conjoint/choice analysis: essential background
4. Practical applications
5. Individual-level vs. aggregate analysis
6. Overfitting and conjoint analysis
7. A two-week part-worth estimation program
8. Early papers
9. Why did HB work for ratings-based conjoint?
10. Improvement for ratings-based conjoint
11. Analysis of choice data
12. IIA
13. Example of IIA
14. IIA and levels of aggregation
15. How does HB reduce IIA problems? (1)
16. How does HB reduce IIA problems? (2)
17. What about interaction effects?
18. What about distributions?
19. Why did HB work for choice-based conjoint?
20. Revolutionary benefit for choice-based conjoint
21. Draws
22. Useful methods of market simulation
23. Draws vs. RFC
24. Why do we care about tuning for scale
25. Shrinkage
26. Problems with shrinkage
27. Covariates in HB
28. Monotonicity constraints (utility constraints)
29. Operationalizing constraints
30. To constrain or not to constrain
31. Uninformative priors?
32. Coding categorical attributes
33. Problems in the posteriors
34. Solutions
35. Dummy coding
36. Effects coding
37. Thank you

Topics Covered

• Introduction to conjoint/choice analysis
• How HB has improved ratings-based conjoint methods
• Why HB has particularly benefited choice-based models
• Market simulation issues including IIA
• Practical considerations such as utility constraints and using proper priors

Links

Series:

• Bayesian Analysis in Marketing

Categories:

• Marketing & Sales

Talk Citation

Publication History

• Published on January 27, 2010