×

\(p\)-values as expert votes. (Les \(p\)-values comme votes d’experts.) (English) Zbl 1061.62510

Summary: The \(p\)-values are often implicitly used as a measure of evidence for hypotheses of tests. This practice has been analyzed with different approaches. It is generally accepted for the one-sided hypothesis problem, but it is often criticized for the two-sided hypothesis problem. We analyze this practice with a new approach to statistical inference. First we select good decision rules without using a loss function, we call them experts. Then we define a probability distribution on the space of experts. The measure of evidence for a hypothesis is the inductive probability of experts that decide this hypothesis.

MSC:

62C99 Statistical decision theory
62C05 General considerations in statistical decision theory
PDFBibTeX XMLCite
Full Text: DOI arXiv EuDML

References:

[1] J.R. Barra, Notions fondamentales de statistique mathématique. Dunod, Paris ( 1971). Zbl0257.62004 MR402992 · Zbl 0257.62004
[2] J.O. Berger, Statistical decision theory and Bayesian analysis, second edition. Springer-Verlag, New York ( 1985). Zbl0572.62008 MR804611 · Zbl 0572.62008
[3] J.O. Berger et T. Sellke, Testing a point null hypothesis: The irreconcilability of p-values and evidence. J. Amer. Statist. Assoc. 82 ( 1987) 112-122. Zbl0612.62022 MR883340 · Zbl 0612.62022 · doi:10.2307/2289131
[4] R.J. Buehler, Fiducial inference, An appreciation, edited by R.A. Fisher. Springer-Verlag, New York, Lecture Notes in Statistics ( 1980) 109-118. MR578886
[5] G. Casella et R.L. Berger, Reconciling Bayesian and frequentist evidence in the one-sided testing problem. J. Amer. Statist. Assoc. 82 ( 1987) 106-111. Zbl0612.62021 MR883339 · Zbl 0612.62021 · doi:10.2307/2289130
[6] J.M. Dickey, Three multidimensional-integral identities with bayesian applications. Ann. Math. Statist. 39 ( 1968) 1615-1628. Zbl0169.50505 MR234545 · Zbl 0169.50505 · doi:10.1214/aoms/1177698417
[7] R.A. Fisher, The fiducial argument in statistical inference. Annals of Eugenics 6 ( 1935) 391-398. JFM62.1345.02 · JFM 62.1345.02
[8] K.R. Gabriel, Simultaneous test procedures - some theory of multiple comparisons. Ann. Math. Statist. 40 ( 1969) 224-250. Zbl0198.23602 MR240931 · Zbl 0198.23602 · doi:10.1214/aoms/1177697819
[9] H.M.J Hung, R.T. O’Neill, P. Bauer et K. Köhne, The behavior of the p-value when the alternative hypothesis is true. Biometrics 53 ( 1997) 11-22. Zbl0876.62015 MR1450180 · Zbl 0876.62015 · doi:10.2307/2533093
[10] J. T. Hwang, G. Casella, C. Robert, M. T. Wells et R.H. Farell, Estimation of accuracy in testing. Ann. Statist. 20 ( 1922) 490-509. Zbl0761.62022 MR1150356 · Zbl 0761.62022 · doi:10.1214/aos/1176348534
[11] S. Karlin, Decision theory for Pólya type distributions. Case of two actions, I, in Proc. Third Berkeley Symposium on Math. Statist. and Prob. Univ. of Calif. Press 1 ( 1955) 115-128. Zbl0073.14202 MR84917 · Zbl 0073.14202
[12] A.H. Kroese, E.A. van der Meulen, K. Poortema et W. Schaafsma, Distributional inference. Statistica Neerlandica 49 ( 1995) 63-82. Zbl0845.62002 MR1333179 · Zbl 0845.62002 · doi:10.1111/j.1467-9574.1995.tb01455.x
[13] E.L. Lehmann, Testing stastical hypotheses, second edition. Wiley, New York ( 1986). Zbl0608.62020 MR852406 · Zbl 0608.62020
[14] E.A. van der Meulen et W. Schaafsma, Assessing weights of evidence for discussing classical statistical hypotheses. Statistics & Decisions 11 ( 1993) 201-220. Zbl0804.62022 MR1257857 · Zbl 0804.62022
[15] A. Monfort, Cours de statistiques mathématique. Économica, Paris ( 1982).
[16] G. Morel, Expertises : procédures statistiques d’aide à la décision, Pré-publication. LAST-Université de Tours ( 1997) 182.
[17] G. Morel, Probabiliser l’espace des décisions, Pub. Sém. 97 : Bru Huber Prum, Paris V ( 1998) 71-98.
[18] C. Robert, L’analyse statistique bayésienne. Économia, Paris ( 1992). Zbl0827.62006 · Zbl 0827.62006
[19] D. Salomé, Statistical inference via fiducial methods, Ph. D. Thesis, Rijksuniversiteit Groningen ( 1998).
[20] W. Schaafsma, J. Tolboom et B. van der Meulen, Discussing truth and falsity by computing a Q-value, in Statistical Data Analysis and Inference, edited by Y. Dodge. North-Holland, Amsterdam ( 1989) 85-100. Zbl0735.62002 MR1089626 · Zbl 0735.62002
[21] H. Scheffé, The analysis of variance, sixième édition. Wiley, New York ( 1970). Zbl0998.62500 MR116429 · Zbl 0998.62500
[22] M.J. Schervish, P-values: What they are and what they are not. Amer. Statist. 50 ( 1996) 203-206. MR1422069
[23] P. Thompson, Improving the admissibility screen: Evaluating test statistics on the basis of p-values rather than power. Comm. Statist. Theory Methods 25 ( 1996) 537-553. Zbl0875.62043 MR1381057 · Zbl 0875.62043 · doi:10.1080/03610929608831712
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.