×

On an asymptotic relative efficiency concept based on expected volumes of confidence regions. (English) Zbl 1434.62102

Summary: The paper deals with an asymptotic relative efficiency concept for confidence regions of multidimensional parameters that is based on the expected volumes of the confidence regions. Under standard conditions the asymptotic relative efficiencies of confidence regions are seen to be certain powers of the ratio of the limits of the expected volumes. These limits are explicitly derived for confidence regions associated with certain plugin estimators, likelihood ratio tests and Wald tests. Under regularity conditions, the asymptotic relative efficiency of each of these procedures with respect to each one of its competitors is equal to 1. The results are applied to multivariate normal distributions and multinomial distributions in a fairly general setting.

MSC:

62H15 Hypothesis testing in multivariate analysis
62F25 Parametric tolerance and confidence regions
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Serfling, R., Approximation theorems of mathematical statistics (1980), New York: Wiley, New York · Zbl 0538.62002
[2] Nikitin, Y., Asymptotic efficiency of nonparametric tests (1995), Cambridge: Cambridge University Press, Cambridge · Zbl 0879.62045
[3] Neyman, J., Outline of a theory of statistical estimation based on the classical theory of probability, Phil Trans Roy Soc, 236, 333-380 (1937) · Zbl 0017.12403 · doi:10.1098/rsta.1937.0005
[4] Lehmann, E.; Romano, J., Testing statistical hypotheses (2005), New York: Wiley, New York · Zbl 1076.62018
[5] Pratt, J., Length of confidence intervals, J Amer Statist Assoc, 56, 549-567 (1961) · Zbl 0099.14002 · doi:10.1080/01621459.1961.10480644
[6] Baringhaus, L.; Gaigall, D., Hotelling’s \(####\) tests in paired and independent survey samples – an efficiency comparison, J Multivariate Anal, 154, 177-198 (2017) · Zbl 1352.62085 · doi:10.1016/j.jmva.2016.11.004
[7] Baringhaus, L.; Gaigall, D., Efficiency comparison of the Wilcoxon tests in paired and independent survey samples, Metrika, 81, 891-930 (2018) · doi:10.1007/s00184-018-0661-4
[8] Bryan, B, Macmahan, Hb, Schafer, Cm, et al. Efficiently computing minimax expected-size confidence regions. Proceedings of the 24th international conference on machine learning; 2007; Corvallis; p. 97-104
[9] Chi-Rong, Li; Tsai-Yu, Lin., Generalized simultaneous confidence regions for regression coefficients and their ratios, J Stat Comput Simul, 87, 1940-1950 (2017) · Zbl 07192041 · doi:10.1080/00949655.2017.1299735
[10] Evans, Sn; Hansen, Bb; Stark, Pb., Minimax expected measure confidence sets for restricted location parameters, Bernoulli, 11, 571-590 (2005) · Zbl 1092.62040 · doi:10.3150/bj/1126126761
[11] Schafer, Cm; Stark, Pb., Constructing confidence regions of optimal expected size, J Am Statist Assoc, 104, 1080-1089 (2009) · Zbl 1388.62065 · doi:10.1198/jasa.2009.tm07420
[12] Wilks, Ss., Shortest average confidence intervals from large samples, Ann Math Statist, 9, 166-175 (1938) · Zbl 0019.35704 · doi:10.1214/aoms/1177732308
[13] Wald, A., Asymptotically shortest confidence intervals, Ann Math Statist, 13, 127-137 (1942) · Zbl 0060.30515 · doi:10.1214/aoms/1177731601
[14] Wilks, Ss; Daly, Jf., An optimum property of confidence regions associated with the likelihood function, Ann Math Statist, 10, 225-235 (1939) · JFM 65.0598.02 · doi:10.1214/aoms/1177732181
[15] Jurečková, J, Confidence sets and intervals. In: Ghosh BK, Sen PK, editor. Handbook of sequential analysis, 269-281. New York: Marcel Dekker; 1991 · Zbl 0753.62046
[16] Baringhaus, L., Asymptotic optimality of multivariate linear hypothesis tests, J Multivariate Anal, 23, 303-311 (1987) · Zbl 0636.62044 · doi:10.1016/0047-259X(87)90159-X
[17] Brown, Ld., Non-local asymptotic optimality of appropriate likelihood ratio tests, Ann Math Statist, 42, 1206-1240 (1971) · Zbl 0247.62008 · doi:10.1214/aoms/1177693236
[18] Hsieh, Hk., On asymptotic optimality of likelihood ratio tests for multivariate normal distributions, Ann Statist, 7, 592-598 (1979) · Zbl 0408.62049 · doi:10.1214/aos/1176344680
[19] Kallenberg, Wcm., Asymptotic optimality of likelihood ratio tests in exponential families (1978), Amsterdam: Mathematisch Centrum, Amsterdam · Zbl 0497.62028
[20] Kourouklis, S., Hodges-Lehmann efficacies of certain tests in multivariate analysis and regression analysis, Canad J Statist, 16, 87-95 (1988) · Zbl 0645.62032
[21] Wald, A., Tests of statistical hypotheses concerning several parameters when the number of observations is large, Trans Amer Math Soc, 54, 426-482 (1943) · Zbl 0063.08120 · doi:10.1090/S0002-9947-1943-0012401-3
[22] Witting, H.; Müller-Funk, U., Mathematische statistik II - Asymptotische statistik: parametrische modelle und nicht-parametrische funktionale (1995), Stuttgart: Teubner, Stuttgart · Zbl 0838.62001
[23] Pratt, J., On interchanging limits and integrals, Ann Math Statist, 31, 74-77 (1960) · Zbl 0090.26802 · doi:10.1214/aoms/1177705988
[24] Königsberger, K., Analysis 2 (2004), Berlin: Springer, Berlin · Zbl 1097.26501
[25] Wilkinson, Jh., The algebraic eigenvalue problem (1965), Oxford: Clarendon Press, Oxford · Zbl 0258.65037
[26] Bahadur, Rr., On Fisher’s bound for asymptotic variances, Ann Math Statist, 55, 1545-1552 (1964) · Zbl 0218.62034 · doi:10.1214/aoms/1177700378
[27] Muirhead, Rj., Aspects of multivariate statistical theory (2005), New York: Wiley, New York
[28] Seber, Gaf., Multivariate observations (1984), New York: Wiley, New York · Zbl 0627.62052
[29] Richard, Jf., A note on the information matrix of the multivariate normal distribution, J Econometrics, 3, 57-60 (1975) · Zbl 0315.62015 · doi:10.1016/0304-4076(75)90066-4
[30] Van Der Vaart, A.; Wellner, J., Weak convergence and empirical processes (1996), New York: Springer, New York · Zbl 0862.60002
[31] Hoeffding, W., Asymptotically optimal tests for multinomial distributions, Ann Math Statist, 36, 369-401 (1965) · Zbl 0135.19706 · doi:10.1214/aoms/1177700150
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.