Paired Comparison Analysis of the van Baaren Model Using Bayesian Approach with Noninformative Prior

Saima Altaf, Muhammad Aslam, Muhammad Aslam

Abstract


One technique being commonly studied these days because of its attractive applications for the comparison of several objects is the method of paired comparisons. This technique permits the ranking of the objects by means of a score, which reflects the merit of the items on a linear scale. The present study is concerned with the Bayesian analysis of a paired comparison model, namely the van Baaren model VI using noninformative uniform prior. For this purpose, the joint posterior distribution for the parameters of the model, their marginal distributions, posterior estimates (means and modes), the posterior probabilities for comparing the two treatment parameters and the predictive probabilities are obtained.

Keywords


Bayesian hypothesis testing; Noninformative priors; Posterior distribution; Predictive probability.

Full Text:

PDF


DOI: http://dx.doi.org/10.18187/pjsor.v8i2.237

Refbacks

  • There are currently no refbacks.




Copyright (c)

Title

Paired Comparison Analysis of the van Baaren Model Using Bayesian Approach with Noninformative Prior

Keywords

Bayesian hypothesis testing; Noninformative priors; Posterior distribution; Predictive probability.

Description

One technique being commonly studied these days because of its attractive applications for the comparison of several objects is the method of paired comparisons. This technique permits the ranking of the objects by means of a score, which reflects the merit of the items on a linear scale. The present study is concerned with the Bayesian analysis of a paired comparison model, namely the van Baaren model VI using noninformative uniform prior. For this purpose, the joint posterior distribution for the parameters of the model, their marginal distributions, posterior estimates (means and modes), the posterior probabilities for comparing the two treatment parameters and the predictive probabilities are obtained.

Date

2012-03-28

Identifier


Source

Pakistan Journal of Statistics and Operation Research; Vol 8. No. 2, 2012



Print ISSN: 1816-2711 | Electronic ISSN: 2220-5810