Bivariate extreme value theory: Models and estimation

SUMMARY Bivariate extreme value distributions arise as the limiting distributions of renormalized componentwise maxima. No natural parametric family exists for the dependence between the marginal distributions, but there are considerable restrictions on the dependence structure. We consider modelling the dependence function with parametric models, for which two new models are presented. Tests for independence, and discriminating between models, are also given. The estimation procedure, and the flexibility of the new models, are illustrated with an application to sea level data. Extreme value theory has recently been an area of much theoretical and practical work. Univariate theory is a well documented area, whereas bivariate/multivariate extreme value theory has, until recently, received surprisingly little attention. In the multivariate case, no natural parametric family exists for the dependence structure, so this must be modelled in some way. In the analysis of environmental extreme value data, there is a need for models of dependence between extremes from different sources: for example at various sea ports, or at various -points of a river. In this paper we consider bivariate extreme value distributions. We assume, without loss of generality, that we have exponential marginal distributions with unit means. The class of bivariate exponential distributions in which we are interested, satisfy a strong stability relation. Exponential variables (X, Y) satisfy the stability relation if and only if W = min (aX, bY) is also exponentially distributed for all a, b > 0 (Pickands, 1981). Therefore, the models we will consider have particular application in reliability and survival analysis. One approach to modelling the dependence structure is via parametric models. This requires a flexible family of models which satisfy certain constraints. Models are of two kinds: either differentiable, or nondifferentiable. All nondifferentiable models give distri- butions which are singular, with nonzero probability concentrated on a certain subspace. The differentiable models have densities, but the existing models are symmetric which leads to the variables being exchangeable. Here, we present two new asymmetric differenti- able models, which have increased flexibility. Properties of the differentiable models are examined. Estimation of the parametric models has previously been by ad hoc methods, because there is a nonregular estimation problem when the margins are independent. For the

[1]  Paul Deheuvels,et al.  Point processes and multivariate extreme values , 1983 .

[2]  J. Tiago de Oliveira.  Bivariate Models for Extremes; Statistical Decision , 1984 .

[3]  Martin Crowder,et al.  A Multivariate Distribution with Weibull Connections , 1989 .

[4]  David R. Cox,et al.  Further Results on Tests of Separate Families of Hypotheses , 1962 .

[5]  S. Resnick,et al.  Limit theory for multivariate sample extremes , 1977 .

[6]  D. Cox,et al.  Parameter Orthogonality and Approximate Conditional Inference , 1987 .

[7]  Probabilistic Aspects of Multivariate Extremes , 1984 .

[8]  R. Fisher,et al.  Limiting forms of the frequency distribution of the largest or smallest member of a sample , 1928, Mathematical Proceedings of the Cambridge Philosophical Society.

[9]  Statistical decision for extremes , 1975 .

[10]  E. J. Gumbel,et al.  Some Analytical Properties of Bivariate Extremal Distributions , 1967 .

[11]  K. V. Mardia,et al.  Asymptotic Independence of Bivariate Extremes , 1964 .

[12]  L.F.M. deHaan,et al.  Statistics of extreme values , 1989 .

[13]  E. J. Gumbel,et al.  Analysis of Empirical Bivariate Extremal Distributions , 1964 .

[14]  B. Efron,et al.  Assessing the accuracy of the maximum likelihood estimator: Observed versus expected Fisher information , 1978 .

[15]  E. Lehmann.  Some Concepts of Dependence , 1966 .

[16]  R. Davies.  Hypothesis testing when a nuisance parameter is present only under the alternative , 1977 .

[17]  William Feller,et al.  An Introduction to Probability Theory and Its Applications , 1967 .

[18]  A. Walden,et al.  Maximum likelihood estimation of the parameters of the generalized extreme-value distribution , 1980 .

[19]  E. Gumbel.  Bivariate Exponential Distributions , 1960 .

[20]  P. Prescott,et al.  Maximum likeiihood estimation of the parameters of the three-parameter generalized extreme-value distribution from censored samples , 1983 .

[21]  D. Cox.  Tests of Separate Families of Hypotheses , 1961 .

[22]  J. Aitchison,et al.  Maximum-Likelihood Estimation of Parameters Subject to Restraints , 1958 .

[23]  M. Woodroofe.  Maximum Likelihood Estimation of a Translation Parameter of a Truncated Distribution , 1972 .

[24]  Henry W. Block,et al.  A Continuous, Bivariate Exponential Extension , 1974 .

[25]  Masaaki Sibuya,et al.  Bivariate extreme statistics, I , 1960 .

[26]  E. Posner,et al.  Application of an Estimator of High Efficiency in Bivariate Extreme Value Theory , 1969 .

[27]  P. Hougaard.  A class of multivanate failure time distributions , 1986 .

[28]  I. Olkin,et al.  A Multivariate Exponential Distribution , 1967 .

[29]  J. D. T. Oliveira,et al.  The Asymptotic Theory of Extreme Order Statistics , 1979 .

[30]  P. Hall.  Fast rates of convergence in the central limit theorem , 1983 .

[31]  P. A. P. Moran,et al.  Maximum-likelihood estimation in non-standard conditions , 1971, Mathematical Proceedings of the Cambridge Philosophical Society.

[32]  Richard L. Smith.  Maximum likelihood estimation in a class of nonregular cases , 1985 .