ArticleOriginal scientific text
Title
Confidence intervals for large non-centrality parameters
Authors 1, 2, 1
Affiliations
- FCT New University of Lisbon, Department of Mathematics, Portugal
- University of Évora, Department of Mathematics, Center for Research on Mathematics and its Applications, Portugal
Abstract
We use asymptotic linearity to derive confidence intervals for large non-centrality parameters. These results enable us to measure relevance of effects and interactions in multifactors models when we get highly statistically significant the values of F tests statistics. We show how to use our approach by considering two sets of data as application examples.
Keywords
asymptotic linearity, non-centrality parameters, highly significant, F tests, measure relevance
Bibliography
- D. Ferreira, S.S. Ferreira, C. Nunes and S. Inácio, Inducing pivot variables and non-centrality parameters in elliptical distributions, AIP Conf. Proc. 1558 (2013), 833.
- J.T. Mexia, Assymptotic Chi-squared Tests, Design and Log-Linear Models (Trabalhos de Investigaçăo, 1. Departamento de Matemática, Faculdade de Cięncias e Tecnologia, Universidade Nova de Lisboa, 1992).
- J.T. Mexia and M.M. Oliveira, Asymptotic linearity and limit distributions, approximations, Journal of Statistical Planning and Inference 140 (2011), 353-357.
- M.M. Oliveira and J.T. Mexia, ANOVA like analysis of matched series of studies with a common structure, Journal of Statistical Planning and Inference 137 (2007), 1862-1870.
- C. Nunes, S.S. Ferreira and J.T. Mexia, Fixed effects NOVA: an extension to samples with random size, Journal of Statistical Computation and Simulation 84 (2014), 2316-2328.
- H. Scheffé, The Analysis of Variance (New York-John Wiley & Sons, 1959).
Pages:
45-56
Main language of publication
English
Received
2015-03-15
Published
2015
DOI: 10.7151/dmps.1175
Exact and natural sciences