This article proposes an alternative approach to incorporate information from observed data with its corresponding prior information using a recipe developed for combining confidence distributions. The outcome function is called a CD posterior, an alternative to Bayes posterior, which is shown here to have the same coverage property as the Bayes posterior. This approach to incorporating a prior distribution has a great advantage that it does not require any prior on nuisance parameters. It also can ease the computational burden which a typical Bayesian analysis endures. An error bound is established on the CD-posterior when there is an error in prior specification.
Primary Subjects: 62A01, 62F03, 62F12, 62F15, 62F40, 62G05, 62G10, 62G20
Bickel D.R. (2006) Incorporation expert knowledge into frequentist inference by combining generalized confidence distributions. Unpublished manuscript.
Boos, D.D. and Monahan, J.F. (1986). Bootstrap methods using prior information. Biometrika, 73, 77–83.
Efron, B. (1986). Why isn’t everybody a Bayesian? The American Statistician, 40(1), 1–5.
Mathematical Reviews (MathSciNet):
Efron, B. (1993). Bayes and likelihood calculations from confidence intervals. Biometrika, 80, 3–26.
Efron, B. (1998). Fisher in 21st Century (with discussion) Stat. Scie., 13, 95–122.
Fraser, D.A.S. (1991). Statistical inference: Likelihood to significance. Journal of the American Statistical Association, 86, 258–265.
Joseph, L., du Berger R., and Belisle P. (1997). Bayesian and mixed Bayesian/likelihood criteria for sample size determination. Statistics in Medicine, 16, 769–781.
Liu, R.Y. and Singh, K. (1993). A quality index based on data-depth and a multivariate rank test. Journal of the American Statistical Association, 88, 257–260.
Liu, R.Y., Parelius, J. and Singh, K. (1999). Multivariate analysis by data-depth: Descriptive statistics, graphics and inference. Ann. Stat., 27, 783–856 (with discussions).
Schweder, T. and Hjort, N.L. (2002). Confidence and Likelihood. Scan. J. Statist., 29, 309–332.
Singh, K., Xie, M. and Strawderman, W.E. (2005). Combining information from independent sources through confidence distribution. Ann. Stat., 33, 159–183.
Singh, K., Xie, M. and Strawderman, W.E. (2007). Confidence distributions - Distribution estimator of a parameter. in Complex Datasets and Inverse Problems. IMS Lecture Notes-Monograph Series, No. 54, (R. Liu, et al., Eds.), 132–150. Festschrift in memory of Yehuda Vardi. IMS Lecture Notes Series.
Wasserman, L. (2007). Why isn’t everyone a Bayesian? in The Science of Bradley Efron. (C.R., Morris and R. Tibshirani, Eds.), 260–261. Springer.
Xie, M., Singh, K. and Strawderman, W.E. (2009). Confidence distributions and a unifying framework for meta-analysis. Technical Report. Department of Statistics, Rutgers University. Submitted for publication.
Xie, M., Liu, R.Y., Damaraju, C.V. and Olson, W.H. (2009). Incorporating expert opinions in the analysis of binomial clinical trials. Technical Report. Department of Statistics, Rutgers University. Submitted for publication.