A Bayesian Model of Sample Selection with a Discrete Outcome Variable: Detecting Depression in Older Adults
Maksym Obrizan ()
No 41, Discussion Papers from Kyiv School of Economics
Depression as a major mental illness among older adults has attracted a lot of research attention. However, the problem of sample selection, inevitable in most health surveys, has been largely ignored. To fill in this gap, this paper formally models selection into the sample jointly with a discrete outcome variable for depression. A Bayesian model of sample selection is developed from a multivariate probit by (i) allowing missing depression status for nonselected respondents, and (ii) using Cholesky factorization of the inverse variance matrix to avoid a Metropolis-Hastings step in the Gibbs sampler. Non-selected respondents are less likely to suffer from depression.
Keywords: Multivariate probit model; Sample selection; Bayesian methods; Gibbs sampler (search for similar items in EconPapers)
JEL-codes: C11 C35 I1 (search for similar items in EconPapers)
New Economics Papers: this item is included in nep-ecm
Note: Journal of Applied Econometrics
References: View references in EconPapers View complete reference list from CitEc
Citations Track citations by RSS feed
Downloads: (external link)
http://repec.kse.org.ua/pdf/KSE_dp41.pdf July 2011 (application/pdf)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: http://EconPapers.repec.org/RePEc:kse:dpaper:41
Access Statistics for this paper
More papers in Discussion Papers from Kyiv School of Economics Contact information at EDIRC.
Series data maintained by Iryna Sobetska ().