Abstract
Given a statistical modelP = {Pθ : θ ∈ x} and a surjective functiong: ϑ→Λ the problem of transformingP into a new modelQ= {λ : λ ∈ Λ} indexed by Λ is investigated. Two characterizations are given for those modelsQ of the form Qλ = ∫ Pθ πλ(dθ), where πλ is some probability such that πλ(g=λ)=1. The first is related to a geometric property ofQ, while the second rests on the inferential implications of adoptingQ. Also, in the first πλ is allowed to be finitely additive, while in the second πλ is σ-additive. Finally, integrated likelihoods are revisited in light of the second characterization.
Similar content being viewed by others
References
Basu, D. (1977). On the elimination of nuisance parameters.Journal of American Statistical Association,72, 355–366.
Berger, J.O., B. Liseo and D. Wolpert (1998). Integrated likelihood methods for eliminating nuisance parameters.Statistical Science, to appear.
Blackwell, D. (1955). On a class of probability spaces.Proceedings of the 3rd Berkeley Symposium in Mathematical Statistics and Probability. University of California Press, 1–6.
Dellacherie, C. and P.A. Meyer (1982).Probabilities and Potential B. North-Holland, Mathematics Studies 72.
Heath, D. and W. Sudderth (1978). On finitely additive priors, coherence, and extended admissibility.Annals of Statistics,6, 333–345.
Srivastava, S.M. (1998).A Course on Borel Sets. Springer Verlag, New York.
Zehna, P.W. (1966). Invariance of maximum likelihood estimators.Annals of Mathematical Statistics,37, 744.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Berti, P., Fattorini, L. & Rigo, P. Eliminating nuisance parameters: two characterizations. Test 9, 133–148 (2000). https://doi.org/10.1007/BF02595855
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02595855
Key Words
- Bayesian inference
- finite additivity
- integrated likelihood
- measurable selection
- nuisance parameter
- statistical model