Simulation algorithm of bayesian approach for choice-conjoint model
Generally in Choice-Conjoint method the Multinomial Logit Model (MNL) is normally used to analyze choice conjoint data, but the MNL has some serious limitations. One of these limitations is the probability to select an alternative over a second alternative must be independent so MNL is not suitable...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Subjects: | |
| Online Access: | http://umt-ir.umt.edu.my:8080/jspui/bitstream/123456789/855/1/T%2057.62%20.Z8%202009%20Abstract.pdf http://umt-ir.umt.edu.my:8080/jspui/bitstream/123456789/855/2/T%2057.62%20.Z8%202009%20FullText.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Generally in Choice-Conjoint method the Multinomial Logit Model (MNL) is normally used to analyze choice conjoint data, but the MNL has some serious limitations. One of these limitations is the probability to select an alternative over a second alternative must be independent so MNL is not suitable for dependent observations is exist in chosen the preferred product or service. As we know, the Multinomial Probit Model (MPM) is a method which assumes that chosen observations are independent but according to researchers the MPM is rarely used due to computational difficulties in computing the maximum likelihood estimates (MLE) for estimate MPM parameters. Therefore this research propose simulation algorithm
of Bayesian approach for estimating parameter in MPM by Bayesian analysis to avoid computational difficulties in computing the maximum likelihood estimates (MLE). |
|---|