Summary
Predictive sample reuse methods usually applied in low structure aparametric paradigms are shown to be useful in certain high structure situations when conjoined with a Bayesian approach. Particular attention is focused on the incomplete data situation for which two alternative sample reuse approaches are devised. The first involves differential weighting and the second a recursive sample reuse algorithm. There are applied to censored exponential survival data. The algorithmic approach appears to be preferable from both a computational and modelling viewpoint.
Similar content being viewed by others
References
Geisser, S. (1971) The inferential use of predictive distributions.Foundations of Statistical Inference. (B.P. Godambe and D.A. Sprott, eds.) 456–69. Toronto, Montreal: Holt, Rinehard and Winston.
— (1974) A predictive approach to the random effect model.Biometrika 61, 101–107.
— (1975a) The predictive sample reuse method with application.J. Amer. Statist. Assoc. 70, 320–328.
— (1975b) Bayesianism, predictive sample reuse, pseudo-observations, and survival.Bull. Internat. Statist. Inst. 40, 285–289.
— (1975c) A new approach to the fundamental problem of applied statistics.Sankhya 37, B 385–397.
Geisser, S. (1976). Predictivism and sample reuse.Proc. 21st Design of Experiments Conference ARO Report 76-2, 385–397.
— (1980) A predictivistic primer.Bayesian Analysis in Econometrics and Statistics: Essays in Honor of Harold Jeffreys. 363–381 (A. Zellner, ed) Amsterdam: North Holland.
Gnedenko, B.V., Belyayev, Yu.K. andSolovyev, A.D. (1969)Mathematical Methods of Reliability Theory. New York: Academic Press.
Stone, M. (1974) Cross-validatory choice and assessment of statistical predictions (with discussion),J. Roy. Statist. Soc. B,36, 111–147
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Geisser, S. Predictive sample reuse techniques for censored data. Trabajos de Estadistica Y de Investigacion Operativa 31, 433–468 (1980). https://doi.org/10.1007/BF02888363
Issue Date:
DOI: https://doi.org/10.1007/BF02888363