Limit theorems for multivariate discrete distributions

According to the usual law of small numbers a multivariate Poisson distribution is derived by defining an appropriate model for multivariate Binomial distributions and examining their behaviour for large numbers of trials and small probabilities of marginal and simultaneous successes. The weak limit law is a generalization of Poisson's distribution to larger finite dimensions with arbitrary dependence structure. Compounding this multivariate Poisson distribution by a Gamma distribution results in a multivariate Pascal distribution which is again asymptotically multivariate Poisson. These Pascal distributions contain a class of multivariate geometric distributions. Finally the bivariate Binomial distribution is shown to be the limit law of appropriate bivariate hypergeometric distributions. Proving the limit theorems mentioned here as well as understanding the corresponding limit distributions becomes feasible by using probability generating functions.     

Sampling distributions associated with the multivariate t distribution

The known sampling distributions and simulation methods associated with multivariate t distributions are reviewed. We believe that this review will serve as an important reference and encourage further research activities in the area.     

The relationship between moments of truncated and original distributions plus some other simple structural properties of weighted distributions

A simple relationship between expectations of positive functions of random variables from truncated and original distributions is pointed out. This is a special case of a relationship between moments of weighted and original distributions which is also discussed. Another topic is that of expressing weighted mixture distributions as mixtures of weighted distributions and vice versa. Finally, an error in some structural properties of weighted distributions given in the literature is identified and corrected. Research carried out while the author held a British Science and Engineering Research Council Postdoctoral Research Fellowship in the Department of Statistics, University of Birmingham, U.K.     

Multivariate t and beta distributions associated with the multivariate F distribution

Relationships between F, skew t and beta distributions in the univariate case are in this paper extended in a natural way to the multivariate case. The result is two new distributions: a multivariate t/skew t distribution (on ℜ^m) and a multivariate beta distribution (on (0,1)^m). A special case of the former distribution is a new multivariate symmetric t distribution. The new distributions have a natural relationship to the standard multivariate F distribution (on (ℜ⁺)^m) and many of their properties run in parallel. We look at: joint distributions, mathematically and graphically; marginal and conditional distributions; moments; correlations; local dependence; and some limiting cases. Received: March 2001     

Community income distributions in a metropolitan area

We extend de Bartolome and Ross [C.A.M. de Bartolome, S.L. Ross, Equilibrium with local governments and commuting: Income sorting vs. income mixing, Journal of Urban Economics 54 (2003) 1–20] to the case when the income distribution in the metropolitan area is a continuous distribution. In particular, we consider a circular central city surrounded by a suburban community. All households must commute to the metropolitan center and public service levels differ in the two jurisdictions. There is intra-jurisdictional and inter-jurisdictional capitalization. Our model has an equilibrium in which the income distributions of the central city and of the suburban community do overlap. Our finding contrasts with the traditional finding of Alonso–Mills–Muth-type models of spatial sorting and of Tiebout-type models of fiscal sorting, both of which have been shown to predict that the income distributions of the two communities do not overlap. In addition, the model explains the fixedness in jurisdictional boundaries.     

On discrete α-unimodal distributions

Some characterizations of a -unimodal lattice distributions, introduced in Abouammoh (1987,19881, and a -monotone lattice distributions, introduced in Steutel (1988), are discussed. Convolution and symmetrization of a -monotone lattice distributions are also investigated. The characterizations and convolution properties obtained by Abouammoh in this connection are corrected and improved.     

Uneven landscapes and city size distributions

This paper proposes a new model generating city size distributions that asymptotically follow the log-normal distribution. The log-normal distribution is consistent with Zipf’s law in the top tail, which is known to hold for many countries in different periods. The key feature of our model is that it can express city size as a product of multiple random factors (e.g., climate, geographic features, and industry composition). Each factor alone need not generate Zipf’s law. Our model provides a justification for classical urban economics models that have been criticized for not delivering Zipf’s law, since a single model typically represents only one factor among many present in reality.     

On some properties of the linear function Poisson binomial and negative binomial distributions

Summary A general model in fluctuations of sums of random variables leading, under certain assumptions, to each of the generalized and linear function Poisson, binomial and negative binomial distributions is presented. Moreover the generating functions and the factorial moments of the linear function Poisson, binomial and negative binomial distributions are obtained in close forms and certain distributional properties are discussed.     

Bounds for the difference between median and mean of beta and negative binomial distributions

Summary We obtain bounds for the difference between the median and mean of the beta and negative binomial distributions using elementary methods.     

Slow convergence of the number of near-maxima for Burr XII distributions

A near-maximum is an observation which falls within a distance a of the maximum observation in an independent and identically distributed sample of size n. Subject to some conditions on the tail thickness of the population distribution, the number K _n (a) of near-maxima is known to converge in probability to one or infinity, or in distribution to a shifted geometric law. In this paper we show that for all Burr XII distributions K _n (a) converges almost surely to unity, but this convergence property may not become clear under certain cases even for very large n. We explore the reason of such slow convergence by studying a distributional continuity between Burr XII and Weibull distributions. We have also given a theoretical explanation of slow convergence of K _n (a) for the Burr XII distributions by showing that the rate of convergence in terms of P{K _n (a) > 1} tending to zero changes very little with the sample size n. Illustrations of the limiting behaviour K _n (a) for the Burr XII and the Weibull distributions are given by simulations and real data. The study also raises an important issue that although the Burr XII provides overall better fit to a given data set than the Weibull distribution, cautions should be taken for the extrapolation of the upper tail behaviour in the case of slow convergence.      

A unified approach to bivariate discrete distributions

Here we introduce a bivariate generalized hypergeometric factorial moment distribution (BGHFMD) through its probability generating function (p.g.f.) whose marginal distributions are the generalized hypergeometric factorial moment distributions introduced by Kemp and Kemp (Bull Int Stat Inst 43:336–338,1969). Well-known bivariate versions of distributions such as binomial, negative binomial and Poisson are special cases of this distribution. A genesis of the distribution and explicit closed form expressions for the probability mass function of the BGHFMD, its factorial moments and the p.g.f.’s of its conditional distributions are derived here. Certain recurrence relations for probabilities, moments and factorial moments of the bivariate distribution are also established.     

Conditional distributions of generalized order statistics and some characterizations

Generalized order statistics have been introduced in Kamps (1995a). They enable a unified approach to several models of ordered random variables, e.g. (ordinary) order statistics, record values, sequential order statistics, record values from non-identical distributions. The purpose of this paper is to develop conditional distributions of one generalized order statistic given another and to characterize the underlying continuous distribution by different conditional expectations. Well-known results for ordinary order statistics and record values are extended to generalized order statistics. Received: July 1997     

Parametric bootstrap tests for continuous and discrete distributions

Statistical procedures based on the estimated empirical process are well known for testing goodness of fit to parametric distribution families. These methods usually are not distribution free, so that the asymptotic critical values of test statistics depend on unknown parameters. This difficulty may be overcome by the utilization of parametric bootstrap procedures. The aim of this paper is to prove a weak approximation theorem for the bootstrapped estimated empirical process under very general conditions, which allow both the most important continuous and discrete distribution families, along with most parameter estimation methods. The emphasis is on families of discrete distributions, and simulation results for families of negative binomial distributions are also presented.     

Forecasting dynamic return distributions based on ordered binary choice

We present a simple approach to the forecasting of conditional probability distributions of asset returns. We work with a parsimonious specification of ordered binary choice regressions that imposes a connection on sign predictability across different quantiles. The model forecasts the future conditional probability distributions of returns quite precisely when using a past indicator and a past volatility proxy as predictors. The direct benefits of the model are revealed in an empirical application to the 29 most liquid U.S. stocks. The forecast probability distribution is translated to significant economic gains in a simple trading strategy. Our approach can also be useful in many other applications in which conditional distribution forecasts are desired.     

More on connections between Wishart and matrix GIG distributions

The paper is devoted to relations between the matrix GIG and Wishart distributions. Our basic tool in the first part is a version of the Matsumoto-Yor property for matrix variables. This approach covers the following issues: the Herz identity for the Bessel function of matrix variate argument, characterization of a class of Wishart matrices and linear transformations of the matrix GIG distribution. The Bayesian Wishart model, studied in the second part, gives an alternative definition of the matrix GIG distribution. Such a model is characterized by linearity of conditional expectations and matrix GIG conditional distribution. It is also extended to Bayesian matrix GIG models, in the framework of which an interesting independence property is proved.     

Tales from tails: On the empirical distributions of forecasting errors and their implication to risk

When evaluating the performances of time series extrapolation methods, both researchers and practitioners typically focus on the average or median performance according to some specific error metric, such as the absolute error or the absolute percentage error. However, from a risk-assessment point of view, it is far more important to evaluate the distributions of such errors, and especially their tails. For instance, a lack of normality and symmetry in error distributions can have significant implications for decision making, such as in stock control. Moreover, frequently these distributions can only be constructed empirically, as they may be the result of a computationally-intensive non-parametric approach, such as an artificial neural network. This study proposes an approach for evaluating the empirical distributions of forecasting methods and uses it to assess eleven popular time series extrapolation approaches across two different datasets (M3 and ForeDeCk). The results highlight some very interesting tales from the tails.     

Determining the optimal process means under mixture normal distributions

In this paper, we consider that the working environment has certain states, and in every state, the parameters of quality characteristics are different. Thus, if we set the characteristics parameters in a specified state, these parameters will change to another state. To describe this situation, we use a mixture of normal distributions, which comprise a flexible and powerful statistical-based modeling tool in practice. Under the step loss function and the piecewise linear loss function, we select the optimal means for the proposed manufacturing process.     

On the modeling of size distributions when technologies are complex

The study considers a stochastic R&D process where the invented production technologies consist of a large number $n$ of complementary components. The degree of complementarity is captured by the elasticity of substitution of the CES aggregator function. Drawing from the Central Limit Theorem and the Extreme Value Theory we find, under very general assumptions, that the cross-sectional distributions of technological productivity are well-approximated either by the lognormal, Weibull, or a novel “CES/Normal” distribution, depending on the underlying elasticity of substitution between technology components. We find the tail of the “CES/Normal” distribution to be fatter than the Weibull tail but qualitatively thinner than the Pareto (power law) one. We also numerically assess the rate of convergence of the true technological productivity distribution to the theoretical limit with $n$ as fast in the body but slow in the tail.     

Matricvariate and matrix multivariate T distributions and associated distributions

Several distributions are studied, simultaneously in the real, complex, quaternion and octonion cases. Specifically, these are the central, nonsingular matricvariate and matrix multivariate T and beta type II distributions and the joint density of the singular values are obtained for real normed division algebras.     

Exact confidence sets and goodness-of-fit methods for stable distributions

Usual inference methods for stable distributions are typically based on limit distributions. But asymptotic approximations can easily be unreliable in such cases, for standard regularity conditions may not apply or may hold only weakly. This paper proposes finite-sample tests and confidence sets for tail thickness and asymmetry parameters (α$\alpha$ and β$\beta$) of stable distributions. The confidence sets are built by inverting exact goodness-of-fit tests for hypotheses which assign specific values to these parameters. We propose extensions of the Kolmogorov–Smirnov, Shapiro–Wilk and Filliben criteria, as well as the quantile-based statistics proposed by McCulloch (1986) in order to better capture tail behavior. The suggested criteria compare empirical goodness-of-fit or quantile-based measures with their hypothesized values. Since the distributions involved are quite complex and non-standard, the relevant hypothetical measures are approximated by simulation, and p$p$-values are obtained using Monte Carlo (MC) test techniques. The properties of the proposed procedures are investigated by simulation. In contrast with conventional wisdom, we find reliable results with sample sizes as small as 25. The proposed methodology is applied to daily electricity price data in the US over the period 2001–2006. The results show clearly that heavy kurtosis and asymmetry are prevalent in these series.