Subject - Statistics Paper - Probability I Module - Distribution Functions II

Total Page:16

File Type:pdf, Size:1020Kb

Subject - Statistics Paper - Probability I Module - Distribution Functions II Subject - Statistics Paper - Probability I Module - Distribution Functions II Distribution functions were defined in the previous module. In this module, we consider a few more of their properties. We begin with an example. Example 1 Consider the distribution function 8 0; if x < 0 <> F (x) = 1 − p; if 0 ≤ x < 1 :>1; if x ≥ 1: For definiteness, take p = 0:6. The figure below shows the graph of F (x) against x. As evident, there is a jump discontinuity at x = 0, of magnitude 0:4, and another one at x = 1, of magnitude 0:6. ♦ We make a few comments on jumps: (i) If a distribution function F has exactly one jump (of magnitude 1) at x = c (a real constant), we say F is degenerate at c, or F is the point mass at c. If a random variable X has this F as its distribution function, then PfX = cg = 1. 1 (ii) F can have either finite or countably infinite number of jump discontinuities. Example 2 A random variable X, defined on some probability space (Ω; A; P), has the Geometric distribution, with parameter θ (2 (0; 1)), if its probability mass func- tion is given by ( θ(1 − θ)x; if x = 0; 1; 2;::: PfX = xg = 0; otherwise. The corresponding distribution function is 80; if x < 0 > > >θ; if 0 ≤ x < 1 > >θ + θ(1 − θ); if 1 ≤ x < 2 > <>θ + θ(1 − θ) + θ(1 − θ)2; if 2 ≤ x < 3 F (x) = fX ≤ xg = X P . >. >n−1 > P x n > θ(1 − θ) = 1 − (1 − θ) ; if n − 1 ≤ x < n >x=0 >. :>. Clearly, FX has countably infinite number of jumps, one at each non-negative integer. ♦ Definition 1 The Lebesgue-Stieltjes measure µF, induced by a continuous distribu- tion function F , is the unique probability measure defined, on the Borel σ-field BR, by µF (a; b] := F (b) − F (a); (L-S) for all bounded left-open right-closed subintervals (a; b] of R. Some explanation of this definition is warranted: (iii) The collection of bounded left-open right-closed subintervals of R, P := f(a; b]: −∞ < a < b < 1g, is a π-system generating BR, and hence, even though µF is defined only for the bounded intervals, it is still completely determined by F through (L-S). (iv) In defining the Lebesgue-Stieltjes measure for continuous distribution func- tions, we have been restrictive. Indeed, the definition applies to discrete distri- bution functions, as well. In this case, the domain of µF need not be the Borel 2 R σ-field BR; the power class 2 can serve the purpose. For continuous distri- bution functions, however, µF (as defined by (L-S)) cannot measure arbitrary subsets of R, and the restriction to BR (or something similar) is necessary. Example 3 Let X be the point mass at c (2 R). Then the distribution function of X is ( 0; if x < c F (x) = 1; if x ≥ c: Define the induced measure µF =: δc by ( 1; if S 3 c δc(S) := 0; if S 63 c; for sets S. The measure δc is often called the (Dirac) delta measure concentrated at c. [While not relevant to the point in hand, we state that if S is kept fixed and c is thought to be the argument, then δc(S) = 1S(c) −− the indicator of S.] ♦ Definition 2 A distribution function F is said to be singular if there exists a linear Borel set B of Lebesgue measure 0 such that µF (B) = 1, where µF is the Lebesgue- Stieltjes measure induced by F . Singular functions form an interesting class of functions, and we discuss a few of their properties: (v) A good look at the definition will reveal that all discrete distributions are singu- lar: the Geometric distribution of Example 2 concentrates all its mass on the set of non-negative integers N[f0g, and it is well-known that1 Leb(N[f0g) = 0; similarly, the point mass at c (Example 3), assumes, with probability 1, the value c, and Lebfcg = 0. The real objects of attention are the continuous singular distribution functions, such as the Cantor function (cf. Distribution Functions I [Example 2]). The following absolutely continuous distribution function 1 1 F (x) = + arctan x; 8x 2 ; 2 π R is, clearly, not singular. (vi) An equivalent characterization of singularity of functions is that a function is singular if and only if its derivative is zero almost everywhere. This fact can be tested on the functions mentioned in the last remark. 1 Leb is Lebesgue measure on BR. 3 Theorem 1 Every distribution function F can be written in the form F = π1Fd + π2Fac + π3Fs; 3 P where each πj ≥ 0 and πj = 1, and Fd;Fac;Fs are, respectively, a discrete, an j=1 absolutely continuous, a continuous singular distribution function. Furthermore, such a decomposition is unique. We do not prove Theorem 1, but instead consider the simpler Theorem 2. For that, we need the following important Lemma. Lemma 1 The set of discontinuity points of a distribution function F is at most countable. Proof: 1 Consider the open intervals of the form (`; ` + 1], for ` 2 Z. Let x1; x2; : : : ; xn be any n points of discontinuity of F in (`; ` + 1], such that F has 1 a jump of magnitude exceeding m (for some positive integer m) at each xj. If the xj's are ordered (i.e.; x1 < x2 < ··· < xn), then, by the non-decreasing property of F , F (`) ≤ F (x1−0) < F (x1) ≤ F (x2−0) < F (x2) ≤ · · · ≤ F (xn−0) < F (xn) ≤ F (`+1): Let pk be the jump at xk; then pk = F (xk)−F (xk −0), for k = 1; 2; : : : ; n. Therefore, n n X X pk = F (xk) − F (xk − 0) ≤ F (` + 1) − F (`): k=1 k=1 1 Since, pk ≥ m ; 8k = 1; 2; : : : ; n, we have n ≤ F (` + 1) − F (`) m () n ≤ m · F (` + 1) − F (`) ≤ m: Even if we consider all points of discontinuity in (`; `+1], and not just a finite number of them, the above inequality remains unchanged. Thus, each interval of unit length has only a finite number of jumps for each m. As m assumes the values 1; 2;::: , we see that the set of discontinuity points in (`; ` + 1] is at most countable. Note that [ R = (`; ` + 1]; `2Z so, the set of all points of discontinuity of F must be at most countable. 4 Theorem 2 Every distribution function F can be written in the form F = κ1Fd + κ2Fc; P where each κj ≥ 0 and κj = 1, Fd is as before, and Fc is a continuous (though j=1;2 not necessarily, absolutely continuous) distribution function. Moreover, this decom- position is unique. Proof: 2 By Lemma, the set of discontinuity points of F is at most countable: let, r1; r2;::: be the points of discontinuity. Let p(rk) := F (rk) − F (rk − 0), and define X Ed(x) := p(rk); for x 2 R; fk:rk≤xg and Ec(x) := F (x) − Ed(x). These definitions imply Ed is a step function and Ec is continuous everywhere on R. Moreover, the functions Ed and Ec satisfy all the properties of a distribution function, except that Ed(1) = 1 and Ec(1) = 1. Ed(1) = 1 iff all the mass of F is concentrated at the jumps, in which case F is discrete and we have F = 1 · Ed + 0. Similarly, Ec(1) = 1 iff F has no jumps at all, and F is continuous with F = 0 + 1 · Ec. For all other situations, 0 < Ed(1);Ec(1) < 1. Define, for all real x, Ed(x) Ec(x) Fd(x) := and Fc(x) := : Ed(1) Ec(1) def These 2 functions are distribution functions. Note that Ed(x) + Ec(x) = F (x); 8x 2 def R, and Ed(1) + Ec(1) = F (1) = 1. Therefore, setting κ1 = Ed(1) and κ2 = Ec(1), we have F = κ1Fd + κ2Fc: Now we prove uniqueness of this decomposition. It suffices to prove that Ed and Ec are uniquely determined, since these, inturn, uniquely determine the corresponding ∗ ∗ distributions. Let Ed + Ec be another decomposition of F . Then, ∗ ∗ ∗ ∗ Ed + Ec = Ed + Ec () Ec − Ec = Ed − Ed : The difference between two continuous functions is continuous, and the difference ∗ between two step functions is, again, a step function. Hence, the equality Ec − Ec = ∗ ∗ ∗ Ed −Ed presents a contradiction. To avoid this, we must have Ec = Ec and Ed = Ed , and thus, the decomposition is unique. 5.
Recommended publications
  • A Test for Singularity 1
    STATISTIII & PROBABIUTY Lk'TTERS ELSEVIER Statistics & Probability Letters 40 (1998) 215-226 A test for singularity 1 Attila Frigyesi *, Ola H6ssjer Institute for Matematisk Statistik, Lunds University, Box 118, 22100 Lund, Sweden Received March 1997; received in revised form February 1998 Abstract The LP-norm and other functionals of the kernel estimate of density functions (as functions of the bandwidth) are studied as means to test singularity. Given an absolutely continuous distribution function, the LP-norm, and the other studied functionals, will converge under some mild side-constraints as the bandwidth decreases. In the singular case, the functionals will diverge. These properties may be used to test whether data comes from an arbitrarily large class of absolutely continuous distributions or not. (~) 1998 Elsevier Science B.V. All rights reserved AMS classification." primary 60E05; secondary 28A99; 62G07 Keywords: Singularity; Absolute continuity; Kernel density estimation 1. Introduction In this paper we aim to test whether a given set of data (= {XI ..... X,}), with a marginal distribution #, stems from an absolutely continuous distribution function or from a singular one. A reference to this problem is by Donoho (1988). Consider a probability measure # on (g~P,~(~P)), where ~(~P) is the Borel a-algebra on R p. The measure tt is said to be singular with respect to the Lebesgue measure 2 if there exists a measurable set A such that /~(AC)=0 and 2(A)=0. If we only require #(A)>0 and 2(A)=0, then /~ is said to have a singular part. If tt does not have a singular part, then it is absolutely continuous with respect to 2.
    [Show full text]
  • Vector Spaces
    ApPENDIX A Vector Spaces This appendix reviews some of the basic definitions and properties of vectm spaces. It is presumed that, with the possible exception of Theorem A.14, all of the material presented here is familiar to the reader. Definition A.I. A set.A c Rn is a vector space if, for any x, y, Z E .A and scalars a, 13, operations of vector addition and scalar multiplication are defined such that: (1) (x + y) + Z = x + (y + z). (2) x + y = y + x. (3) There exists a vector 0 E.A such that x + 0 = x = 0 + x. (4) For any XE.A there exists y = -x such that x + y = 0 = y + x. (5) a(x + y) = ax + ay. (6) (a + f3)x = ax + f3x. (7) (af3)x = a(f3x). (8) 1· x = x. Definition A.2. Let .A be a vector space and let JV be a set with JV c .A. JV is a subspace of .A if and only if JV is a vector space. Theorem A.3. Let .A be a vector space and let JV be a nonempty subset of .A. If JV is closed under addition and scalar multiplication, then JV is a subspace of .A. Theorem A.4. Let .A be a vector space and let Xl' ••• , Xr be in .A. The set of all linear combinations of Xl' .•• ,xr is a subspace of .A. Appendix A. Vector Spaces 325 Definition A.S. The set of all linear combinations of Xl'OO',X, is called the space spanned by Xl' ..
    [Show full text]
  • Quantization of Self-Similar Probability Measures and Optimal Quantizers by Do˘Ganc¸¨Omez Department of Mathematics, North Dakota State University
    Quantization of self-similar probability measures and optimal quantizers by Do˘ganC¸¨omez Department of Mathematics, North Dakota State University Outline: 1. Five-minute Crash Course in Probability 2. Probability Measures and Distributions 3. Voronoi Partitions 4. Centers and Moments of Probability Distributions 5. Quantization 6. Some Recent Results n Throughout we will deal with R with a norm k k: In particular k kr will be the r-norm for some 1 ≤ r < 1: Some of the information below are adopted from Foundations of Quantization for Probability Distributions, by S. Graf and H. Luschgy, Lecture Notes in Mathematics, # 1730, Springer, 2000. 1. Five-minute Crash Course in Probability. (Ω; F;P ) : a probability space. A measurable function X :Ω ! R is called a random variable. R The expectation (or mean) of X is defined as E(X) := Ω XdP: (i) E(Xk): k-th moment of X; and E(jXjk): k-th absolute moment of X: (ii) E[(X − E(X))k]: k-th central moment of X; and E[jX − E(X)jk]: k-th absolute central moment of X: Note: If k = 1;E[(X − E(X))k] = 0: When k = 2;E[(X − E(X))2] is called the variance of X: If X is a random variable, then it induces a probability measure on Borel subsets of R by −1 PX := P ◦ X : (Sometimes PX is called a law.) The distribution (function) of a r.v. X is a function F : R ! R such that F (x) = PX (−∞; x] = P (f! : X(!) ≤ xg): Observe that F (b) − F (a) = PX (a; b]: Fact.
    [Show full text]
  • THE DERIVATION of the CHI-Square TEST of GOODNESS of FIT
    THE DERIVATION OF THE CHI-sQUARE TEST OF GOODNESS OF FIT Jacqueline Prillaman Submitted to McGill University in partial fulfilment of the requirements for the de6ree of Master of Science, August, 1956 I wish to express my gratitude to Prof. W. Kozakiewicz for his generous assistance and valuable guidance in the preparation of this thesis. TABLE OF CONTENTS Introduction .................................................. 1 Part One: An Intuitive Approach to the Chi-Square Test of Goodness of Fit ............................ 5 I: Presentation of the Test II: The Number of Degrees of Fre?dom III: Greenhood's Derivation of the Chi-Square Test of Goodness of Fit Part Two: A Rigorous De velopment of the Chi- Square Test of Goodness of Fit .......................... 22 I: The Normal Distribution in the Space of n-Dimensions II: The Limiting Distribution of the Chi-Square Variable by the Method of Characteristic Functions Introduction Let ~, ~, . ., x be s independent sample values drawn from 5 a population of normally distributed values with zero mean and unit variance. Then the variable 2 2 2 u = x1 + x 2 + . + xs is said to follow the ~t distribution with parameter s. This para­ meter is called the number of degrees of freedom for reasons which will be explained later. Using the moment-generating function we shall prove that s s-2 u 2 2- (1) u e 2 u > o. f(u) = r(~) (!) Since the basic variable, x, in the population has the frequency function f(x) = m.- e , we have, for G < l2 , M 2(&) = Mx2(&) = xi +• 2 ~ (1-2&) = 1 e -2 dx rh 1-· 2 1 .
    [Show full text]
  • On Lin's Condition for Products of Random Variables with Singular Joint
    PROBABILITY AND MATHEMATICAL STATISTICS Vol. 40, Fasc. 1 (2020), pp. 97–104 Published online 2.4.2020 doi:10.37190/0208-4147.40.1.6 ON LIN’S CONDITION FOR PRODUCTS OF RANDOM VARIABLES WITH SINGULAR JOINT DISTRIBUTION BY ALEXANDER I L’ I N S K I I (KHARKOV) AND SOFIYA O S T ROV S K A (ANKARA) Abstract. Lin’s condition is used to establish the moment determi- nacy/indeterminacy of absolutely continuous probability distributions. Re- cently, a number of papers related to Lin’s condition for functions of random variables have appeared. In the present paper, this condition is studied for products of random variables with given densities in the case when their joint distribution is singular. It is proved, assuming that the densities of both random variables satisfy Lin’s condition, that the density of their product may or may not satisfy this condition. 2020 Mathematics Subject Classification: Primary 60E05; Secondary 44A60. Key words and phrases: random variable, singular distribution, Lin’s con- dition. 1. INTRODUCTION Lin’s condition plays a significant role in the investigation of the moment deter- minacy of absolutely continuous probability distributions. Generally speaking, it is used along with Krein’s logarithmic integral to establish the so-called ‘converse criteria’. See, for example, [4, Section 5, Theorems 5–10] and [7, Section 2]. This condition was introduced and applied by G. D. Lin [3], while the name ‘Lin’s con- dition’ was put forth in [10]. Let us recall the pertinent notions. DEFINITION 1.1. Let f be a probability density continuously differentiable on (0; +∞).
    [Show full text]
  • Multivariate Stable Distributions
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Elsevier - Publisher Connector JOURNAL OF MULTIVARIATE ANALYSIS 2, 444-462 (1972) Multivariate Stable Distributions S. J. PRESS* Graduate School of Business, University of Chicago, Chicago, Illinois 60637 Communicated by A. P. Dempster This paper is devoted to the theory and application of multidimensional stable distributions. Properties of these laws are developed and explicit algebraic representations are given in terms of characteristic functions. Symmetric and asymmetric laws and their properties are studied separately. A measure of association for variables following a symmetric bivariate stable distribution is provided and its properties are shown to be analogous to those of the ordinary correlation coefficient. Examples are given, and the class of symmetric multi- variate stable distributions is applied to a problem in portfolio analysis. 1. INTRODUCTION This work is concerned with probabilistic aspects of multivariate stable distributions. In Section 2, this class of distributions is defined, algebraic characteristic function representations are given, and some properties of the classare presented. Section 3 defines the symmetric multivariate stable distribu- tions and presents some of their properties. Since the variables under study generally do not possesssecond moments, correlation coefficients do not exist. However, an extended notion of correlation coefficient applicable to symmetric multivariate stable laws is presented in Section
    [Show full text]
  • Statistical Theory Distribution Theory Reading in Casella and Berger
    STAT 450: Statistical Theory Distribution Theory Reading in Casella and Berger: Ch 2 Sec 1, Ch 4 Sec 1, Ch 4 Sec 6. Basic Problem: Start with assumptions about f or CDF of random vector X = (X1;:::;Xp). Define Y = g(X1;:::;Xp) to be some function of X (usually some statistic of interest like a sample mean or correlation coefficient or a test statistic). How can we compute the distribution or CDF or density of Y ? Univariate Techniques In a univariate problem we have X a real-valued random variable and Y , a real valued transformation of X. Often we start with X and are told either the density or CDF of X and then asked to find the distribution of something like X2 or sin(X) { a function of X which we call a transformation. I am going to show you the steps you should follow in an example problem. Method 1: compute the CDF by integration and differentiate to find fY . Example: Suppose U ∼ Uniform[0; 1]. What is the distribution of − log(X)?. Step 1: Make sure you give the transformed variable a name. Let Y = − log U. Step 2: Write down the definition of the CDF of Y : FY (y) = P (Y ≤ y): Step 3: Substitute in the definition of Y in terms of the original variable { in this case, U: FY (y) = P (Y ≤ y) = P (− log(U) ≤ y): Step 4: Solve the inequality to get, hopefully, one or more intervals for U. In this case − log(U) ≤ y is the same as log(U) ≥ −y 1 which is the same as U ≥ e−y: So P (− log(U) ≤ Y ) = P U ≥ e−y : Step 5: now use the assumptions to write the resulting probability in terms of FU .
    [Show full text]
  • Splitting Models for Multivariate Count Data Jean Peyhardi, Pierre Fernique, Jean-Baptiste Durand
    Splitting models for multivariate count data Jean Peyhardi, Pierre Fernique, Jean-Baptiste Durand To cite this version: Jean Peyhardi, Pierre Fernique, Jean-Baptiste Durand. Splitting models for multivariate count data. Journal of Multivariate Analysis, Elsevier, 2021, 181, pp.104677. 10.1016/j.jmva.2020.104677. hal- 02962877 HAL Id: hal-02962877 https://hal.inria.fr/hal-02962877 Submitted on 9 Oct 2020 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Splitting models for multivariate count data Jean Peyhardia,∗, Pierre Ferniqueb, Jean-Baptiste Durandc aIMAG, Universit´ede Montpellier, place Eug`eneBataillon, 34090 Montpellier, France bCIRAD, UMR AGAP, 34098 Montpellier cLaboratoire Jean Kuntzmann, 38058 Grenoble, France Abstract We introduce the class of splitting distributions as the composition of a singular multivariate distribution and an uni- variate distribution. This class encompasses most common multivariate discrete distributions (multinomial, negative multinomial, multivariate hypergeometric, multivariate negative hypergeometric, . ) and contains several new ones. We highlight many probabilistic properties deriving from the compound aspect of splitting distributions and their un- derlying algebraic properties. These simplify the study of their characteristics, inference methods, interpretation and extensions to regression models. Parameter inference and model selection are thus reduced to two separate problems, preserving time and space complexity of the base models.
    [Show full text]
  • the MULTIVARIATE T-DISTRIBUTION ASSOCIATED
    THE MULTIVARIATE t-DISTRIBUTION ASSOCIATED WITH A SET OF NORMAL SAMPLE DEVIATES By E. A. OORNIsn* [Manu8cript received July 29, 1954] Sumnwry This paper gives a short account of the more important properties of the multi­ variate t-distribution, which arises in association with a set of normal sample deviates. I. INTRODUCTION The -multivariate t-distribution described herein was encountered in the course of a recent investigation of the frequency distribution of the spectro­ graphic error which occurs in the D.O. arc excitation of samples of soil and plant ash (Oertel and Oornish 1953). The original observations consisted of triplicate and sextuplicate determinations of copper, manganese, molybdenum, and tin in samples containing varying amounts of each of these elements, and it was found that the variance of measured line intensity was proportional to the square of the mean intensity. A logarithmic transformation stabilized the variance, which, for any individual element, could then be estimated from all the samples involving 1hat element, and by subsequent standardization all the values of the new metri' could be placed on a comparable basis. The standardized variates appeared to be normally distributed, and it became desirable, for various reasons, to test this point. In providing tests of normality, two general lines of approach have been / followed: (1) a normal distribution is fitted to the sample data, and the X2 test of goodness of fit is applied; (2) certain functions of the sample moments are calculated, and the significance of their departure from expectations based on the assumption of normality is examined, but neither of these procedures, as ordinarily used, was suitable owing to the particular nature of the data.
    [Show full text]
  • Some Methods of Constructing Multivariate Distributions Abderrahmane Chakak Iowa State University
    Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations 1993 Some methods of constructing multivariate distributions Abderrahmane Chakak Iowa State University Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Statistics and Probability Commons Recommended Citation Chakak, Abderrahmane, "Some methods of constructing multivariate distributions " (1993). Retrospective Theses and Dissertations. 10411. https://lib.dr.iastate.edu/rtd/10411 This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps.
    [Show full text]
  • Econstor Wirtschaft Leibniz Information Centre Make Your Publications Visible
    A Service of Leibniz-Informationszentrum econstor Wirtschaft Leibniz Information Centre Make Your Publications Visible. zbw for Economics Erdely, Arturo Article Value at risk and the diversification dogma Revista de Métodos Cuantitativos para la Economía y la Empresa Provided in Cooperation with: Universidad Pablo de Olavide, Sevilla Suggested Citation: Erdely, Arturo (2017) : Value at risk and the diversification dogma, Revista de Métodos Cuantitativos para la Economía y la Empresa, ISSN 1886-516X, Universidad Pablo de Olavide, Sevilla, Vol. 24, pp. 209-219 This Version is available at: http://hdl.handle.net/10419/195388 Standard-Nutzungsbedingungen: Terms of use: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Documents in EconStor may be saved and copied for your Zwecken und zum Privatgebrauch gespeichert und kopiert werden. personal and scholarly purposes. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle You are not to copy documents for public or commercial Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich purposes, to exhibit the documents publicly, to make them machen, vertreiben oder anderweitig nutzen. publicly available on the internet, or to distribute or otherwise use the documents in public. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, If the documents have been made available under an Open gelten abweichend von diesen Nutzungsbedingungen die in der dort Content Licence (especially Creative Commons Licences), you genannten Lizenz gewährten Nutzungsrechte. may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-sa/4.0/ www.econstor.eu REVISTA DE METODOS´ CUANTITATIVOS PARA LA ECONOM´IA Y LA EMPRESA (24).
    [Show full text]
  • Doubly Singular Matrix Variate Beta Type I and II and Singular Inverted Matricvariate $ T $ Distributions
    Doubly singular matrix variate beta type I and II and singular inverted matricvariate t distributions Jos´eA. D´ıaz-Garc´ıa ∗ Department of Statistics and Computation 25350 Buenavista, Saltillo, Coahuila, Mexico E-mail: [email protected] Ram´on Guti´errez J´aimez Department of Statistics and O.R. University of Granada Granada 18071, Spain E-mail: [email protected] Abstract In this paper, the densities of the doubly singular beta type I and II distributions are found, and the joint densities of their corresponding nonzero eigenvalues are provided. As a consequence, the density function of a singular inverted matricvariate t distribution is obtained. 1 Introduction Let A and B be independent Wishart matrices or, alternatively, let B = YY′ where Y has a matrix variate normal distribution. Then the matrix variate beta type I distribution is arXiv:0904.2147v1 [math.ST] 14 Apr 2009 defined in the literature as (A + B)−1/2B(A + B)−1/2, U = B1/2(A + B)−1B1/2, (1) ′ −1 Y (A + B) Y. Analogously, the matrix variate beta type II distribution is defined as A−1/2BA−1/2, F = B1/2A−1B1/2, (2) ′ −1 Y A Y. These can be classified as central, noncentral or doubly noncentral, depending on whether A and B are central, B is noncentral or A and B are noncentral, respectively; see D´ıaz-Garc´ıaand Guti´errez-J´aimez ∗Corresponding author Key words. Singular distribution, matricvariate t distribution, matrix variate beta type I distribution, matrix variate beta type II distribution. 2000 Mathematical Subject Classification. 62E15, 15A52 1 (2007, 2008b).
    [Show full text]