¿Por qué la Inferencia Estadística Bayesiana?revisitado
-
1
Universidad Complutense de Madrid
info
ISSN: 1889-3805
Año de publicación: 2020
Volumen: 36
Número: 3
Páginas: 231-237
Tipo: Artículo
Otras publicaciones en: BEIO, Boletín de Estadística e Investigación Operativa
Resumen
The objective of this paper is to update the one already published in the journal Bolet´ın de la Sociedad de Estad´ıstica e Investigaci´on Operativa “in 2006 with title ¿Por qu´e la Inferencia Estad´ıstica Bayesiana? at Bulletin of Statistics and Operations Research (BEIO)”. Section 1 discusses the two main approaches to statistical inference. Section 2 highlights the problems that occur when the accuracy in the frequentist approach is specified for point estimators, confidence intervals and in the hypothesis tests under Neyman-Pearson’s theory. Section 3 introduces the Bayesian approach and Birnbaum’s theorem, which implies that the frequentist approach can violate sufficiency and conditionality. Finally, Section 4 explains how to measure accuracy by using the final distribution.
Referencias bibliográficas
- 1] Bernardo, J. M. (2003) Bayesian Statistics. Encyclopedia of Life Support Systems. Probability and Statistics. (R. Viertl. ed) Oxford: UNESCO.
- [2] Bernardo, J.M. and Smith, A. F. M. (1994) Bayesian Theory. New York: Wiley.
- [3] Birnbaum, A. (1962) On the foundations of statistical inference, Journal of American Statistical Association, 57, 269-326.
- [4] Gómez-Villegas, M. A. (2005, 2008, 2011, 2014, 2016, 2018) Inferencia Estad´ıstica. Madrid: Díaz de Santos.
- [5] Gómez-Villegas, M. A. (2006) ¿Por qué la Inferencia Estad´ıstica Bayesiana? Bolet´ın de la Sociedad de Estad´ıstica e Investigaci´on Operativa, 22, 1, 6-8.
- [6] Jeffreys, H. (1939) Theory of Probability (3ª ed. 1961). Londres: Oxford University Press.