二维正态分布的假设检验

将二维随机向量分解成互不相关的主成分,通过对两主成分的正态独立性检验达到二维随机向量正态性检验的目的.     

Normal correlation coefficient of non-normal variables using piece-wise linear approximation

The correlation coefficient of non-normal variables is expressed as a function of the correlation coefficient of normal variables using piece-wise linear approximation of each univariate transform of normal to anything, and the second order moments of a multiply truncated bivariate normal distribution. For the inverse problem, an algorithm iterates this analytic function in order to assign a normal correlation coefficient to two non-normal variables. The algorithm is applied for the generation of randomized bivariate samples with given correlation coefficient and marginal distributions and used in a randomization test for bivariate nonlinearity. The test correctly does not reject the null hypothesis of linear correlation if the nonlinearity is plausible and due to the sample transform alone.     

UMVU estimation of the ratio of powers of normal generalized variances under correlation

We consider estimation of the ratio of arbitrary powers of two normal generalized variances based on two correlated random samples. First, the result of Iliopoulos [Decision theoretic estimation of the ratio of variances in a bivariate normal distribution, Ann. Inst. Statist. Math. 53 (2001) 436-446] on UMVU estimation of the ratio of variances in a bivariate normal distribution is extended to the case of the ratio of any powers of the two variances. Motivated by these estimators’ forms we derive the UMVU estimator in the multivariate case. We show that it is proportional to the ratio of the corresponding powers of the two sample generalized variances multiplied by a function of the sample canonical correlations. The mean squared errors of the derived UMVU estimator and the maximum likelihood estimator are compared via simulation for some special cases.     

Number of Records in a Bivariate Sample with Application to Missouri River Flood Data

For a sequence of observations from a bivariate absolutely continuous distribution, two types of records are considered depending on whether a univariate record is established in both or in at least one of the components. The distributional properties of the associated univariate and bivariate record indicators are examined. Correlation between the number of component records and the first two moments of the number of bivariate records in a finite random sample are obtained. These are evaluated for the Farlie-Gumbel-Morgenstern and bivariate normal distributions. Large sample properties of these moments are explored. Our results are used to predict the number of record annual floods at two sites along the Missouri river during the next 50 years.     

Constant Local Dependence

The local dependence function is constant for the bivariate normal distribution. Here we identify all other distributions which also have constant local dependence. The key property is exponential family conditional distributions and a linear conditional mean. When given two marginal distributions only, this characterisation is not very helpful, and numerical solutions are necessary.     

Bivariate inverse gaussian distribution

Summary A bivariate inverse Gaussian (IG) density function is constructed. Relations of the bivariate IG distribution to the normal and χ² distributions are established. The corresponding bivariate random walk (RW) density function is obtained. The properties and behaviour of bivariate IG distribution are studied for large parametric values. Moment estimates of the five parameters are given and applications are pointed out. A generalization to the multivariate IG distribution is proposed.     

二元Weinman型指数分布的特征及其应用

导出了Weinman型二元指数分布的一个特征,由此获得了参数θj(j=0,1)的最大似然估计及矩估计,给出了二元Weinman型指数分布的二种模拟,还得到了强度为二元Weinman分布时并联结构系统可靠度的估计.     

Convolutions of multivariate phase-type distributions

This paper is concerned with multivariate phase-type distributions introduced by Assaf et al. (1984). We show that the sum of two independent bivariate vectors each with a bivariate phase-type distribution is again bivariate phase-type and that this is no longer true for higher dimensions. Further, we show that the distribution of the sum over different components of a vector with multivariate phase-type distribution is not necessarily multivariate phase-type either, if the dimension of the components is two or larger.     

A class of bivariate exponential distributions

We introduce a class of absolutely continuous bivariate exponential distributions, generated from quadratic forms of standard multivariate normal variates.This class is quite flexible and tractable, since it is regulated by two parameters only, derived from the matrices of the quadratic forms: the correlation and the correlation of the squares of marginal components. A simple representation of the whole class is given in terms of 4-dimensional matrices. Integral forms allow evaluating the distribution function and the density function in most of the cases.The class is introduced as a subclass of bivariate distributions with chi-square marginals; bounds for the dimension of the generating normal variable are underlined in the general case.Finally, we sketch the extension to the multivariate case.     

Some properties of the bivariate lognormal distribution for reliability applications

In this paper, we study the bivariate lognormal distribution from a reliability point of view. The conditional distribution of X given Y > y is found to be log‐skew normal. The monotonicity of the hazard rates of the univariate as well as the conditional distributions is discussed. Clayton's association measure is obtained in terms of the hazard gradient, and its value in the case of our model is derived. The probability distributions, in the case of series and parallel systems, are derived, and the monotonicity of their failure rates is discussed. Three real applications of the bivariate lognormal distribution are provided, two from financial economics and one from reliability. Copyright © 2012 John Wiley & Sons, Ltd.     

The Sample Covariance Is Not Efficient for Elliptical Distributions

It is well known that the sample covariance is not an efficient estimator of the covariance of a bivariate normal vector. We extend this result to elliptical distributions and we propose a simple explicit estimator, which is efficient in the normal case and which outperforms the sample covariance in general. Necessary and sufficient conditions are established under which this estimator is in general efficient for an elliptical distribution.     

Making Use of Incomplete Observations for Regression in Bivariate Normal Model

Two estimates of the regression coefficient in bivariate normal distribution are considered: the usual one based on a sample and a new one making use of additional observations of one of the variables. They are compared with respect to variance. The same is done for two regression lines. The conclusion is that the additional observations are worth using only when the sample is very small.     

A simple diagonals-parameter symmetry and quasi-symmetry model

Goodman (1979) proposed a class of diagonals-parameter symmetry models for square contingency tables with ordered categories, A simple version of that model is considered in which the log odds parameters have a linear pattern. The model is also a simple quasi-symmetry model. It fits well when there is an underlying bivariate normal distribution.     

Bivariate distributions characterized by one family of conditionals and conditional percentile or mode functions

It is well known that full knowledge of all conditional distributions will typically serve to completely characterize a bivariate distribution. Partial knowledge will often suffice. For example, knowledge of the conditional distribution of X given Y and the conditional mean of Y given X is often adequate to determine the joint distribution of X and Y. In this paper, we investigate the extent to which a conditional percentile function or a conditional mode function (of Y given X), together with knowledge of the conditional distribution of X given Y will determine the joint distribution. Finally, using this methodology a new characterization of the classical bivariate normal distribution is given.     

A note on Sheppard's corrections for grouping and maximum likelihood estimation

Sheppard's corrections for grouping can, in the case of an underlying normal distribution, be interpreted as a first step to the solution of the maximum likelihood equations which incorporate the grouping problem. This result of Lindley (for the univariate) and Haitovsky (for the bivariate) is generalized to the multivariate normal distribution, making use of recent results in matrix algebra. Also, formulae concerning the efficiency lost in grouping are generalized to the multivariate case.     

A measure of independence for a multivariate normal distribution and some connections with factor analysis

This paper gives results for the population value of a measure of the goodness-of-fit of a general multivariate normal distribution to the simpler hypothesis of independent normal variables. The measure was introduced by Rudas, Clogg and Lindsay in 1994, who gave the value for the bivariate normal distribution. Connections with factor analysis are briefly discussed.     

Modeling heterogeneity for bivariate survival data by the compound Poisson distribution with random scale

We propose a bivariate Weibull regression model with heterogeneity (frailty or random effect) which is generated by compound Poisson distribution with random scale. We assume that the bivariate survival data follow bivariate Weibull of Hanagal (2004). There are some interesting situations like survival times in genetic epidemiology, dental implants of patients and twin births (both monozygotic and dizygotic) where genetic behavior (which is unknown and random) of patients follows a known frailty distribution. These are the situations which motivate us to study this particular model. We propose a two stage maximum likelihood estimation procedure for the parameters in the proposed model and develop large sample tests for testing significance of regression parameters.     

二元Block～Basu型指数分布的特征及其应用

导出了二元Block~Basu型指数分布的一个特征,利用该特征,获得了二元Block~Basu型指数分布参数的最大似然估计及矩估计,给出了强度服从二元Block~Basu型分布时并联结构系统可靠度的估计,并给出了二元Block~Basu型指数分布的一个随机模拟.     

The bivariate normal copula function is regularly varying

We derive the rate of decay of the tail dependence of the bivariate normal distribution and establish its link with regularly varying functions. This result is an initial step in explaining the discrepancy between a zero asymptotic tail dependence coefficient and mass in the tail of a joint distribution.     

A multivariate skew normal distribution

In this paper, we define a new class of multivariate skew-normal distributions. Its properties are studied. In particular we derive its density, moment generating function, the first two moments and marginal and conditional distributions. We illustrate the contours of a bivariate density as well as conditional expectations. We also give an extension to construct a general multivariate skew normal distribution.