Bayesian analysis of a multivariate spatial ordered probit model. Issue 4 (3rd April 2023)

• Record Type: Journal Article
• Authors: Gao, Ping
• Is Part Of: Communications in statistics. Volume 52:Issue 4(2023)
• Page Start: 1300
• Page End: 1317
• Publication Date: 2023-04-03
• Subjects: Bayesian inference -- Markov chain Monte Carlo (MCMC) -- Multivariate outcomes -- Ordered probit model -- Spatial dependences; Mathematical statistics -- Periodicals; Mathematical statistics -- Data processing -- Periodicals; Digital computer simulation -- Periodicals; 519.5
• DOI: 10.1080/03610918.2021.1879857
• Languages: English
• ISSNs: 0361-0918

Abstract

Abstract: Many phenomena are correlated and spatially dependent. In addition, some of them are ordinal discrete responses. Thus, a model is needed to capture the interactions of multivariate ordinal outcomes and spatial dependences. Following Smith and LeSage (2004) and Jeliazkov et al. (2008), this study proposes a new algorithm for a multivariate spatial ordered probit (MSOP) model to address this need. In applying this model, the parameters are calculated using the Bayesian inference based on Markov chain Monte Carlo (MCMC) sampling. The validity and accuracy of the MSOP model is verified by simulated datasets, and the model performs very well with the simulated data. In addition, this study illustrates the model by applying it to two response variables, self-rated health, and life satisfaction of elderly people in 18 representative provinces in mainland China. The empirical results show that the spatial dependences are indispensable on the response variables.