Bivariate and multivariate distributions with bimodal marginals (original) (raw)
Related papers
The Meta-elliptical Distributions with Given Marginals
Journal of Multivariate Analysis, 2002
Based on an analysis of copulas of elliptically contoured distributions, joint densities of continuous variables with given strictly increasing marginal distributions are constructed. A method utilized for this procedure is to embed the spherical distribution quantile transformation of each variable into an elliptically contoured distribution. The new class of distributions is then called meta-elliptical distributions. The corresponding analytic forms of the density, conditional distribution functions, and dependence properties are derived. This new class of distributions has the same Kendall's rank correlation coefficient as meta-Gaussian distributions. As an extension of elliptically contoured distributions, some new classes of distributions are also obtained.
On Some Circular Distributions Induced by Inverse Stereographic Projection
Sankhya B, 2019
In earlier studies of circular data, the corresponding probability distributions considered were mostly assumed to be symmetric. However, the assumption of symmetry may not be meaningful for some data. Thus there has been increased interest, more recently, in developing skewed circular distributions. In this article we introduce three skewed circular models based on inverse stereographic projection, originally introduced by Minh and Farnum (2003), by considering three different versions of skewed-t considered in the literature, namely skewed-t by Azzalini (1985), two-piece skewed-t (Fernández and Steel, 1998) and skewedt by Jones and Faddy (2003). Shape properties of the resulting circular distributions along with estimation of parameters using maximum likelihood are also discussed in this article. Further, real data sets are used to illustrate the application of the new models. It is found that Azzalini and Jones-Faddy skewed-t versions are good competitors, however, the Jones-Faddy version is computationally more tractable.
On Elliptical Possibility Distributions
IEEE Transactions on Fuzzy Systems
This paper aims to propose two main contributions in the field of multivariate data analysis through the possibility theory. The first proposition is the definition of a generalized family of multivariate elliptical possibility distributions. These distributions have been derived from a consistent probabilitypossibility transformation over the family of so-called elliptical probability distributions. The second contribution proposed by the paper is the definition of two divergence measures between possibilistic distributions. We prove that a symmetric version of the Kullback-Leibler divergence guarantees all divergence properties when related to the space of possibility distributions. We further derive analytical expressions of the latter divergence and of the Hellinger divergence for certain possibility distributions pertaining to the elliptical family proposed, especially the normal multivariate possibility divergence in two dimensions. Finally, this paper provides an illustration of the developed possibilistic tools in an application of bi-band change detection between optical satellite images.
New Families of Distributions for Modeling Bivariate Data, with Applications
Journal of Mathematics and Statistics
In this study we introduce a new method of adding two shape parameters to any baseline bivariate distribution function (df) to get a more flexible family of bivariate df's. Through the additional parameters we can fully control the type of the resulting family. This method is applied to yield a new two-parameter extension of the bivariate standard normal distribution, denoted by BSSN. The statistical properties of the BSSN family are studied. Moreover, via a mixture of the BSSN family and the standard bivariate logistic df, we get a more capable family, denoted by FBSSN. Theoretically, each of the marginals of the FBSSN contains all the possible types of df's with respect to the signs of skewness and excess kurtosis. In addition, each possesses very wide range of the indices of skewness and kurtosis. Finally, we compare the families BSSN and FBSSN with some important competitors (i.e., some generalized families of bivariate df's) via real data examples.
2006
We propose a new bivariate distribution with five shape parameters and two scale parameters. It allows great flexibility for the bivariate situation and successfully replaces the assumption of the bivariate normal distribution when the ratio of two positive bounded variables is in question
An elliptically symmetric angular Gaussian distribution
We define a distribution on the unit sphere S d−1 called the elliptically symmetric angular Gaussian distribution. This distribution, which to our knowledge has not been studied before, is a subfamily of the angular Gaussian distribution closely analogous to the Kent subfamily of the general Fisher–Bingham distribution. Like the Kent distribution, it has ellipse-like contours, enabling modelling of rotational asymmetry about the mean direction, but it has the additional advantages of being simple and fast to simulate from, and having a density and hence likelihood that is easy and very quick to compute exactly. These advantages are especially beneficial for computationally intensive statistical methods, one example of which is a parametric bootstrap procedure for inference for the directional mean that we describe.
ON SOME NEW CLASSES OF MULTIVARIATE PROBABILITY DISTRIBUTIONS
New Classes of transformations on the Euclidean n-space R n are applied to sets of n (n = 2, 3, …) independent random variables T 1 , …, T n. Basically two cases are considered: 1) the random variables are all Weibullian, and 2) they are all gamma. The transformations, when applied to the random vectors (T 1 , …, T n), produce, as outputs, r. vectors (X 1 , …, X n) whose joint probability densities are investigated. The new probability densities of (X 1 , …, X n) produced by the Weibullian input r. variables T 1 , …, T n are called " pseudoWeibullians " , and the other " pseudogammas ". The origin and the basis for new methods of pdfs construction is set of problems associated with multicomponent system reliability modeling. In this context the random variables X 1 , …, X n are used to represent the system components life times.
An order-statistics-based method for constructing multivariate distributions with fixed marginals
Journal of Multivariate Analysis, 2008
A new system of multivariate distributions with fixed marginal distributions is introduced via the consideration of random variates that are randomly chosen pairs of order statistics of the marginal distributions. The distributions allow arbitrary positive or negative Pearson correlations between pairs of random variates and generalise the Farlie-Gumbel-Morgenstern distribution. It is shown that the copulas of these distributions are special cases of the Bernstein copula. Generation of random numbers from the distributions is described, and formulas for the Kendall and grade (Spearman) correlations are given. Procedures for data fitting are described and illustrated with examples.
A New Class of Distributions Generated by the Extended Bimodal-Normal Distribution
Journal of Probability and Statistics, 2018
In this study, we present a new family of distributions through generalization of the extended bimodal-normal distribution. This family includes several special cases, like the normal, Birnbaum-Saunders, Student’s t, and Laplace distribution, that are developed and defined using stochastic representation. The theoretical properties are derived, and easily implemented Monte Carlo simulation schemes are presented. An inferential study is performed for the Laplace distribution. We end with an illustration of two real data sets.