Improved hypothesis testing in a general multivariate elliptical model. (English) Zbl 07192009
Summary: This paper investigates improved testing inferences under a general multivariate elliptical regression model. The model is very flexible in terms of the specification of the mean vector and the dispersion matrix, and of the choice of the error distribution. The error terms are allowed to follow a multivariate distribution in the class of the elliptical distributions, which has the multivariate normal and Student-\(t\) distributions as special cases. We obtain Skovgaard’s adjusted likelihood ratio (LR) statistics and Barndorff-Nielsen’s adjusted signed LR statistics and we compare the methods through simulations. The simulations suggest that the proposed tests display superior finite sample behaviour as compared to the standard tests. Two applications are presented in order to illustrate the methods.
MSC:
| 62-XX | Statistics |
Keywords:
elliptical model; general parameterization; modified likelihood ratio statistic; modified signed likelihood ratio statistic; multivariate normal distribution; multivariate Student-\(t\) distributionSoftware:
OxReferences:
| [1] | Barndorff-Nielsen OE. Inference on full or partial parameters, based on the standardized signed log likelihood ratio. Biometrika. 1986;73:307-322. [Web of Science ®], [Google Scholar] · Zbl 0605.62020 |
| [2] | Skovgaard IM. Likelihood asymptotics. Scand J Statist. 2001;28:3-32. doi: 10.1111/1467-9469.00223[Crossref], [Web of Science ®], [Google Scholar] · Zbl 0965.62014 |
| [3] | Peña EA, Rohatgi VK, Szekely GJ. On the non-existence of ancillary statistics. Statist Probab Lett. 1992;15:357-360. doi: 10.1016/0167-7152(92)90153-V[Crossref], [Web of Science ®], [Google Scholar] · Zbl 0756.62003 |
| [4] | Lemonte AJ, Patriota AG. Multivariate elliptical models with general parameterization. Stat Methodol. 2011;8:389-400. doi: 10.1016/j.stamet.2011.03.001[Crossref], [Web of Science ®], [Google Scholar] · Zbl 1215.62054 |
| [5] | Verbeke G, Molenberghs G. Linear mixed models for longitudinal data. New York: Springer; 2000. [Google Scholar] · Zbl 0956.62055 |
| [6] | Cheng CL, Van Ness JW. Statistical regression with measurement error. London: Oxford University Press; 1999. [Google Scholar] · Zbl 0947.62046 |
| [7] | Vanegas LH, Paula GA. A semiparametric approach for joint modeling of median and skewness. Test. 2015;24:110-135. doi: 10.1007/s11749-014-0401-7[Crossref], [Web of Science ®], [Google Scholar] · Zbl 1315.62058 |
| [8] | Fang KT, Kotz S, Ng KW. Symmetric multivariate and related distributions. London: Chapman and Hall; 1990. [Crossref], [Google Scholar] · Zbl 0699.62048 |
| [9] | Melo TFN, Ferrari SLP, Patriota AG. Improved estimation in a general multivariate elliptical model; 2015. arXiv:1508.05994. [Google Scholar] |
| [10] | Severini TA. Likelihood methods in statistics. New York: Oxford University Press; 2000. [Google Scholar] · Zbl 0984.62002 |
| [11] | Sen PK, Singer JM. Large sample methods in statistics. An introduction with applications. New York: Chapman & Hall; 1993. [Crossref], [Google Scholar] · Zbl 0867.62003 |
| [12] | Doornik JA. Object-oriented matrix programming using Ox. London: Timberlake Consultants Press; 2013 (ISBN 978-0-9571708-1-0). [Google Scholar] |
| [13] | Pace L, Salvan A. Principles of statistical inference from a neo-Fisherian perspective. London: World Scientific; 1997. [Crossref], [Google Scholar] · Zbl 0911.62003 |
| [14] | Brazzale AR, Davison AC. Accurate parametric inference for small samples. Statist Sci. 2008;23:465-484. doi: 10.1214/08-STS273[Crossref], [Web of Science ®], [Google Scholar] · Zbl 1329.62101 |
| [15] | Cribari-Neto F, Queiroz MPF. On testing inference in beta regressions. J Stat Comput Simul. 2014;84:186-203. doi: 10.1080/00949655.2012.700456[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 1453.62384 |
| [16] | Ferrari SLP, Pinheiro EC. Small-sample likelihood inference in extreme-value regression models. J Stat Comput Simul. 2014;84:582-595. doi: 10.1080/00949655.2012.720686[Taylor & Francis Online], [Web of Science ®], [Google Scholar] · Zbl 1453.62486 |
| [17] | Guolo A. Higher-order likelihood inference in meta-analysis and meta-regression. Stat Med. 2012;31:313-327. doi: 10.1002/sim.4451[Crossref], [PubMed], [Web of Science ®], [Google Scholar] |
| [18] | Lemonte AJ, Ferrari SLP. Signed likelihood ratio tests in the Birnbaum-Saunders regression model. J Statist Plann Inference. 2011;141:1031-1040. doi: 10.1016/j.jspi.2010.09.007[Crossref], [Web of Science ®], [Google Scholar] · Zbl 1200.62124 |
| [19] | Fraser DAS, Reid N, Wu J. A simple general formula for tail probabilities for frequentist and Bayesian inference. Biometrika. 1999;86:249-264. doi: 10.1093/biomet/86.2.249[Crossref], [Web of Science ®], [Google Scholar] · Zbl 0932.62003 |
| [20] | Rosillo FG, Martín N, Egido MA. Comparison of conventional and accelerated lifetime testing of fluorescent lamps. Light Res Technol. 2010;42:243-259. doi: 10.1177/1477153509356019[Crossref], [Web of Science ®], [Google Scholar] |
| [21] | Crepeau H, Koziol J, Reid N, et al. Analysis of incomplete multivariate data from repeated measurement experiments. Biometrics. 1985;41:505-514. doi: 10.2307/2530875[Crossref], [PubMed], [Web of Science ®], [Google Scholar] · Zbl 0614.62086 |
| [22] | Smith SP. Differentiation of the Cholesky algorithm. J Comput Graph Statist. 1995;4:134-147. [Taylor & Francis Online], [Google Scholar] |
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.
