Author
Listed:
- Negrín, Miguel A.
(Departamento de Métodos Cuantitativos e Instituto Universitario TiDES. Universidad de Las Palmas de Gran Canaria, Las Palmas de Gran Canarias (España))
- Martel, María
(Departamento de Métodos Cuantitativos e Instituto Universitario TiDES. Universidad de Las Palmas de Gran Canaria, Las Palmas de Gran Canarias (España))
- Vázquez-Polo, Francisco J.
(Departamento de Métodos Cuantitativos e Instituto Universitario TiDES. Universidad de Las Palmas de Gran Canaria, Las Palmas de Gran Canarias (España))
Abstract
El objetivo de este trabajo es construir un procedimiento bayesiano, fácil de implementar en la práctica, que soslaye los problemas encontrados en los métodos de meta-análisis para el caso de datos binarios con alta presencia de ceros. Para ello, consideraremos el problema dentro de uno general de selección bayesiana de modelos identificando de forma adecuada sus elementos. En particular definiendo una sencilla distribución “link” entre las efectividades por estudios y la meta-efectividad del tratamiento. Hemos desarrollado un procedimiento objetivo bayesiano para las medidas de interés así como para el test de comparación de efectividades entre tratamientos. Presentamos dos aplicaciones con datos reales. Básicamente, probamos en este trabajo que es posible desarrollar un test de hipótesis bayesiano de igualdad de tratamientos sin necesidad de procedimientos ad-hoc artificiales que actúan sobre transformaciones de las variables y observaciones que no siempre son adecuadas. Respecto a los ejemplos, no existen evidencias empíricas para considerar que existen diferencias entre tratamientos. || The propose of this paper is to develop a Bayesian procedure that adequately account for studies with zero observations in meta-analysis and then we focus the problem in the context of the Bayesian selection models. Also, attention is focused to the link distribution between effectiveness in each study/center and the meta-effectiveness. We present an objective Bayesian method where all quantities of interest jointly with a Bayesian test for equality between treatments are also obtained. A couple of examples with are developed in depth using the pro- posed Bayesian meta-analysis for the binomial model. Basically, we obtain a Bayesian model for meta-analysis for sparse binomial data without considering transformations and/or corrections in variable/parameters. In respect to the examples considered, we do not find a relevant difference between treatments.
Suggested Citation
Negrín, Miguel A. & Martel, María & Vázquez-Polo, Francisco J., 2015.
"Una nota sobre un procedimiento bayesiano para meta-análisis con datos binarios con alta presencia de ceros || A note on a Bayesian procedure for meta-analysis of rare data,"
Revista de Métodos Cuantitativos para la Economía y la Empresa = Journal of Quantitative Methods for Economics and Business Administration, Universidad Pablo de Olavide, Department of Quantitative Methods for Economics and Business Administration, vol. 20(1), pages 64-76, December.
Handle:
RePEc:pab:rmcpee:v:20:y:2015:i:1:p:64-76
Download full text from publisher
Corrections
All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:pab:rmcpee:v:20:y:2015:i:1:p:64-76. See general information about how to correct material in RePEc.
If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.
We have no bibliographic references for this item. You can help adding them by using this form .
If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Publicación Digital - UPO (email available below). General contact details of provider: https://edirc.repec.org/data/dmupoes.html .
Please note that corrections may take a couple of weeks to filter through
the various RePEc services.