×

A model for perturbed production or measurement processes involving compound normal distributions. (English) Zbl 0937.90027

Summary: To solve an easy-to-understand real world problem, we invite the reader to travel through many useful concepts related to the important models provided by the compound normal distributions. This class of distributions is a subclass of the multivariate elliptical distributons that allows a clearcut interpretation of a data generating mechanism useful in certain situations. A multivariate elliptical distribution is characterized by the existence of a particular function \(\varphi\). In the special case of a compound normal distribution, the function \(\varphi\) appears to be simply the (one-sided) Laplace-Stieltjes Transform (LST) of the mixing distribution. This fact allows in particular to easily express the covariance matrix and the kurtosis parameter of the compound normal in terms of derivatives, evaluated at the origin, of the LST of the mixing distribution. An example of application of the results is the asymptotic theory for canonical correlation analysis. The asymptotic distributions of the sample canonical correlation coefficients (and of statistics used for testing hypotheses about the population coefficients) have very simple forms in the case of compound normal distributions. They depend on the LST of the mixing distribution through the kurtosis parameter. Here we focus our attention on the inference on the simple correlation coefficient \(\rho\) of the components of bivariate compound normal distributions. We illustrate with a simple example in industrial engineering how inferenec on \(\rho\) is affected by the choice of the associated mixing distribution.

MSC:

90B30 Production models
62H05 Characterization and structure theory for multivariate probability distributions; copulas
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Abramowitz, M.; Stegun, I. A., Handbook of Mathematical Functions, ((1965), Dover: Dover New York), 1020-1029
[2] Chmielewski, M. A., Elliptically symmetric distributions: A review and bibliography, Int. Statist. Rev., 49, 67-74 (1981) · Zbl 0467.62047
[3] Devlin, S. J.; Gnanadesikan, R.; Kettenring, J. R., Some multivariate applications of elliptical distributions, (Ideka, S.; etal., Essays in Probability and Statistics (1976), Shinko Tsusho: Shinko Tsusho Tokyo), 365-395 · Zbl 0364.62015
[4] Doetsch, G., Einfuerung in Theorie und Anwendung der Laplace-Transformation (1958), Birkhauser: Birkhauser Basel
[5] Fang, K. T.; Anderson, T. W., Statistical Inference in Elliptically Contoured and Related Distributions (1990), Allerton Press: Allerton Press New York · Zbl 0747.00016
[6] Fang, K. T.; Kotz, S.; Ng, K. W., Symmetric Multivariate and Related Distributions (1989), Chapman & Hall: Chapman & Hall London
[7] Fang, K. T.; Zhang, Y., Generalized Multivariate Analysis, (Science (1989), Springer: Springer Berlin), Moscow
[8] Forster, O., Analysis III, Integral Rechnung in \(R^n\) mit Anwendungen (1984), Vieweg: Vieweg Braunschweig
[9] Kelker, D., Distribution theory of spherical distributions and a location scale parameter generalization, Sankhia Ser. A, 32, 419-430 (1970) · Zbl 0223.60008
[10] Lange, K. L.; Little, R. J.A.; Taylor, J. M.G., Robust statistical modelling using the \(t\) distribution, J. Amer. Statist. Ass., 408, 881-896 (1989)
[11] Mitchell, A. F.S.; Krzanowski, W. J., The mahalanobis distance and elliptic distributions, Biometrika, 72, 464-467 (1985) · Zbl 0571.62042
[12] Maronna, R. A., Robust M-estimators of multivariate location and scatter, Ann. Statist., 4, 51-67 (1976) · Zbl 0322.62054
[13] Muirhead, R. J., The effects of elliptical distributions on some standard procedures involving correlation coefficients: A review, (Gupta, R. P., Multivariate Statistical Analysis (1980), North-Holland: North-Holland Amsterdam), 143-159
[14] Muirhead, R. J., Aspects of Multivariate Statistical Theory (1982), Wiley: Wiley New York · Zbl 0556.62028
[15] Muirhead, R. J.; Waterneaux, C. M., Asymptotic distributions in canonical correlation analysis and other multivariate procedures for nonnormal populations, Biometrika, 67, 31-43 (1980) · Zbl 0448.62037
[16] Rolle, J. D., Best nonnegative invariant partially orthogonal quadratic estimation in normal regression, J. Amer. Statist. Ass., 428, 1378-1385 (1994) · Zbl 0810.62066
[17] Rolle, J. D., Using aggregated \(p\)-way classified data for the estimation of a dispersion matrix, (Technical Report, 1995.19 (1995), HEC, University of Geneva)
[18] Rolle, J. D., Optimization of functions of matrices with application in statistics and econometrics, Linear Algebra Appl., 234, 261-275 (1996) · Zbl 0841.15012
[19] Widder, D. W., The Laplace Transform (1941), Princeton Univ. Press: Princeton Univ. Press Princeton, NJ · JFM 67.0384.01
[20] Zellner, A., Bayesian and non-Bayesian analysis of the regression model with multivariate student \(t\) error terms, J. Amer. Statist. Ass., 71, 400-405 (1976) · Zbl 0348.62026
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.