DOCSLIB.ORG

Arxiv:2108.13047V1 [Cond-Mat.Stat-Mech] 30 Aug 2021

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Arxiv:2108.13047V1 [Cond-Mat.Stat-Mech] 30 Aug 2021 Measurement induced quantum walks A. Didi and E. Barkai Department of Physics, Institute of Nanotechnology and Advanced Materials, Bar-Ilan University, Ramat-Gan 5290002, Israel (Dated: August 31, 2021) We investigate a tight binding quantum walk on a graph. Repeated stroboscopic measurements of the position of the particle yield a measured "trajectory", and a combination of classical and quantum mechanical properties for the walk are observed. We explore the effects of the measure- ments on the spreading of the packet on a one dimensional line, showing that except for the Zeno limit, the system converges to Gaussian statistics similarly to a classical random walk. A large deviation analysis and an Edgeworth expansion yield quantum corrections to this normal behavior. We then explore the first passage time to a target state using a generating function method, yielding properties like the quantization of the mean first return time. In particular, we study the effects of certain sampling rates which cause remarkable change in the behavior in the system, like divergence of the mean detection time in finite systems and a decomposition of the phase space into mutually exclusive regions, an effect that mimics ergodicity breaking, whose origin here is the destructive interference in quantum mechanics. For a quantum walk on a line we show that in our system the first detection probability decays classically like (time)−3=2, this is dramatically different compared to local measurements which yield a decay rate of (time)−3, indicating that the exponents of the first passage time depends on the type of measurements used. I. INTRODUCTION time τ, at which point we measure it's position. The free evolution between measurements is given by unitary dy- namics determined by the Hamiltonian of the system, Consider a non-biased random walk in dimension one, typically described by the adjacency matrix of a graph. for example a walk with Gaussian increments, or random After measurement is complete we log the location that walk on a lattice with equal probability to jump left or we found the particle in and then allow it to freely evolve right. In the long time limit the density of particles all for τ time once more, before repeating the process, such starting at the origin will converge to a Gaussian distri- as in the case presented in Fig 1. Our goals in this work bution. On the other hand, for a tight binding quantum are twofold, to better understand the statistical proper- walk on a line with nearest neighbor jumps, it is well ties of such a process were it to repeat indefinitely, and to known that the PDF, given by the square modulus of examine the first detection statistics when we are inter- the wave function, will propagate ballistically in the ab- ested in the amount of time it will take for the system to sence of measurement. The question we seek to answer reach a certain state under this measurement protocol. is, what will happen in the case where we measure the An analogous classical example [1] would be to record position of the particle stroboscopically, as explained in the position of an animal randomly wandering in the the abstract? These measurements yield a path that the wilderness at a constant frequency 1/τ. By doing this particle took, given by the sites that it was detected at. many times, we can define a path that the animal took From this path we may construct in principle the prob- as the list of positions we recorded, with this we can also ability of finding the particle on lattice point x after n ask questions about the statistics of how long it would measurements. Will the repeated collapse of the wave take the animal to travel between two different locations, function cause the statistics of the system to display fea- or to return to it's starting place. Where this analogy tures typical of a classical random walk, or will the pro- breaks down however, is that while for an animal the act cess remain a purely quantum one in spite of this, or of recording its position and the frequency of recording some combination of the two? In our work we found that wouldn't change it's path, in our case it is the driving the measurements induce a Gaussian behavior, similar to force of the random walk, as we'll later show. what one would expect of a classical random walk, how- The rest of the paper is organized as follows: In Secs. arXiv:2108.13047v1 [cond-mat.stat-mech] 30 Aug 2021 ever a closer look at the problem using large-deviation II, III we define our model with regard to both the mea- theory and an Edgeworth expansion reveal the quantum surement induced quantum random walk without termi- nature of the process. Notwithstanding these effects, the nation and the first detection problem. In Sec. IV we convergence to the Gaussian behavior is typically fast, describe the mathematical framework we'll be using and with the exception of the Zeno limit τ ! 0, where the in Secs. V, VI we use it to derive several properties of non-classical effects are better pronounced. this type of random walk. In Sec. VII we apply the To more precisely define our model, we examine a aforementioned results and mathematical tools to study a closed quantum system which is prepared in some defi- measurement induced quantum random walk finite graph nite initial state and evolves unitarily for a predetermined in detail and in Sec. VIII we do the same for an infinite 2 ψin〉 tween nearest neighbours on the infinite line with an am- -3-2-1 0 1 2 3 plitude γ for the jumps, and hence for an initial condi- tion set at the origin these dynamics are sometimes called a tight binding quantum walk [9]. In this example the FIG. 1: A segment of the infinite line. For a particle space X is defined with the vertices on the lattice jxi starting at the origin, we allow it to unitarily evolve for and as usual X^ jxi = x jxi. A schematic representation τ time, and then we measure it's position. The of this Hamiltonian is given in Fig. 1. Another example measurement localizes it's wave function to some single is presented in Fig. 2 where we have a finite ring. It site on the lattice, which we record as being the is well known that the statistical properties of first de- location of the particle at time τ. After this we allow tection time for finite and infinite systems dramatically the particle to freely evolve for τ time once more before differ [1, 6, 10] , so we will later use these two models as we measure it's position again. By repeating this examples in sections VII and VIII. process many times, we generate a "path" that the particle took. For the lattice shown here, an example of a path would be: f0; 2; 1; 0; −3; :::g 0 1 1d lattice. We follow this up in Sec. IX with a compari- son of our measurement scheme to a local measurement model, as such models have seen extensive research in 5 2 papers by Dhar et al. [2, 3], Krovi and Brun [4, 5], and others. In particular, we'll focus on directly comparing our results to those found for the scheme considered in [6{8]. We close the paper with discussions and a sum- mary in Sec. X. Detailed calculations and an additional 4 3 example are presented in the appendices. FIG. 2: Schematic model of a benzene ring. The edges represent jumps between nearest neighbours, all with II. MODEL the same amplitude. In Sec VII we examine the properties of a measurement induced quantum walk on Our model can broadly be divided into two primary this graph, as well as the first detection problem. elements. The first are the quantum dynamics described by the Schr¨odingerequation. The second being the mea- surement protocol which yields the well defined meaning In order to establish the position of the particle at for the position of the particle. Beginning with a de- every step of the random walk, we start by defining the scription of the first part, we consider a single particle measurement protocol as follows: Position measurements whose time evolution is described by a time independent are performed at discrete times t = τ; 2τ; :::Nτ using the ^ Hermitian Hamiltonian H according to the Schr¨odinger position operator X. Between each measurement event d the dynamics are unitary as described above. At every equation H j i = i~ j i where we set ~ = 1. The ini- dt measurement we obtain the exact position of the particle tial wave function is denoted j ini. The H is represented here with a graph where nodes describe states and edges causing the wave function to collapse and be localized describe the hopping amplitudes between these states. to that point on the graph. Of course, the basic postu- Our main focus considers a H which is described by some lates of quantum mechanics mean that the outcome of adjacency matrix, though the theory presented below is the measurement is random. Over the course of many more general. measurements this process produces a list of the loca- The Hilbert space we consider is discrete and is tions the particle was detected at. As the measurements spanned by X which is the set of eigenvectors of the op- combined with the unitary time evolution given by the erator X^. X^ can be any arbitrary Hermitian operator Schr¨odingerequation act as the generators of this ran- as long as all of its eigenvalues are non-degenerate.
Load more

Recommended publications
  • Effects of Scale-Dependent Non-Gaussianity on Cosmological
    Preprint typeset in JHEP style - HYPER VERSION Effects of Scale-Dependent Non-Gaussianity on Cosmological Structures Marilena LoVerde1, Amber Miller1, Sarah Shandera1, Licia Verde2 1 Institute of Strings, Cosmology and Astroparticle Physics Physics Department, Columbia University, New York, NY 10027 2 ICREA, Institut de Ci`encies de l’Espai (ICE), IEEC-CSIC, Campus UAB, F. de Ci`encies, Torre C5 par-2, Barcelona 08193, Spain Emails: [email protected], [email protected], [email protected], [email protected] Abstract: The detection of primordial non-Gaussianity could provide a powerful means to test various inflationary scenarios. Although scale-invariant non-Gaussianity (often de- scribed by the fNL formalism) is currently best constrained by the CMB, single-field models with changing sound speed can have strongly scale-dependent non-Gaussianity. Such mod- els could evade the CMB constraints but still have important effects at scales responsible for the formation of cosmological objects such as clusters and galaxies. We compute the effect of scale-dependent primordial non-Gaussianity on cluster number counts as a function of redshift, using a simple ansatz to model scale-dependent features. We forecast constraints arXiv:0711.4126v3 [astro-ph] 13 Mar 2008 on these models achievable with forthcoming data sets. We also examine consequences for the galaxy bispectrum. Our results are relevant for the Dirac-Born-Infeld model of brane inflation, where the scale-dependence of the non-Gaussianity is directly related to the geometry of the extra dimensions. Contents 1. Introduction 1 2. General Single-field Inflation: Review and Example 6 2.1 General Single Field Formalism 6 2.2 Scale-Dependent Relationships 8 2.3 An Example from String Theory 9 3.
    [Show full text]
  • Gaussian and Non-Gaussian-Based Gram-Charlier and Edgeworth Expansions for Correlations of Identical Particles in HBT Interferometry
    Gaussian and non-Gaussian-based Gram-Charlier and Edgeworth expansions for correlations of identical particles in HBT interferometry by Michiel Burger De Kock Thesis presented in partial fulfilment of the requirements for the degree of Master of Theoretical Physics at the University of Stellenbosch Department of Physics University of Stellenbosch Private Bag X1, 7602 Matieland, South Africa Supervisor: Prof H.C. Eggers March 2009 Declaration By submitting this thesis electronically, I declare that the entirety of the work contained therein is my own, original work, that I am the owner of the copyright thereof (unless to the extent explicitly otherwise stated) and that I have not previously in its entirety or in part submitted it for obtaining any qualification. Date: 02-03-2009 Copyright c 2009 Stellenbosch University All rights reserved. i Abstract Gaussian and non-Gaussian-based Gram-Charlier and Edgeworth expansions for correlations of identical particles in HBT interferometry M.B. De Kock Department of Physics University of Stellenbosch Private Bag X1, 7602 Matieland, South Africa Thesis: MSc (Theoretical Physics) March 2009 Hanbury Brown{Twiss interferometry is a correlation technique by which the size and shape of the emission function of identical particles created during collisions of high-energy leptons, hadrons or nuclei can be determined. Accurate experimental datasets of three-dimensional correlation functions in momentum space now exist; these are sometimes almost Gaussian in form, but may also show strong deviations from Gaussian shapes. We investigate the suitability of expressing these correlation functions in terms of statistical quantities beyond the normal Gaussian description. Beyond means and the covariance matrix, higher-order moments and cumulants describe the form and difference between the measured correlation function and a Gaussian distribution.
    [Show full text]
  • On the Validity of the Formal Edgeworth Expansion for Posterior Densities
    On the validity of the formal Edgeworth expansion for posterior densities John E. Kolassa Department of Statistics and Biostatistics Rutgers University [email protected] Todd A. Kuffner Department of Mathematics Washington University in St. Louis [email protected] October 6, 2017 Abstract We consider a fundamental open problem in parametric Bayesian theory, namely the validity of the formal Edgeworth expansion of the posterior density. While the study of valid asymptotic expansions for posterior distributions constitutes a rich literature, the va- lidity of the formal Edgeworth expansion has not been rigorously established. Several authors have claimed connections of various posterior expansions with the classical Edge- worth expansion, or have simply assumed its validity. Our main result settles this open problem. We also prove a lemma concerning the order of posterior cumulants which is of independent interest in Bayesian parametric theory. The most relevant literature is synthe- sized and compared to the newly-derived Edgeworth expansions. Numerical investigations illustrate that our expansion has the behavior expected of an Edgeworth expansion, and that it has better performance than the other existing expansion which was previously claimed to be of Edgeworth-type. Keywords and phrases: Posterior; Edgeworth expansion; Higher-order asymptotics; arXiv:1710.01871v1 [math.ST] 5 Oct 2017 Cumulant expansion. 1 Introduction The Edgeworth series expansion of a density function is a fundamental tool in classical asymp- totic theory for parametric inference. Such expansions are natural refinements to first-order asymptotic Gaussian approximations to large-sample distributions of suitably centered and normalized functionals of sequences of random variables, X1;:::;Xn. Here, n is the avail- able sample size, asymptotic means n ! 1, and first-order means that the approximation 1 using only the standard Gaussian distribution incurs an absolute approximation error of order O(n−1=2).
    [Show full text]
  • New Edgeworth-Type Expansions with Finite Sample Guarantees1]A2
    New Edgeworth-type expansions with finite sample guarantees1 Mayya Zhilova2 School of Mathematics Georgia Institute of Technology Atlanta, GA 30332-0160 USA e-mail: [email protected] Abstract: We establish higher-order expansions for a difference between probability distributions of sums of i.i.d. random vectors in a Euclidean space. The derived bounds are uniform over two classes of sets: the set of all Euclidean balls and the set of all half-spaces. These results allow to account for an impact of higher-order moments or cumulants of the considered distributions; the obtained error terms depend on a sample size and a dimension explicitly. The new inequalities outperform accuracy of the normal approximation in existing Berry–Esseen inequalities under very general conditions. For symmetrically distributed random summands, the obtained results are optimal in terms of the ratio between the dimension and the sample size. The new technique which we developed for establishing nonasymptotic higher-order expansions can be interesting by itself. Using the new higher-order inequalities, we study accuracy of the nonparametric bootstrap approximation and propose a bootstrap score test under possible model misspecification. The results of the paper also include explicit error bounds for general elliptical confidence regions for an expected value of the random summands, and optimality of the Gaussian anti-concentration inequality over the set of all Euclidean balls. MSC2020 subject classifications: Primary 62E17, 62F40; secondary 62F25. Keywords and phrases: Edgeworth series, dependence on dimension, higher-order accuracy, multivariate Berry–Esseen inequality, finite sam- ple inference, anti-concentration inequality, bootstrap, elliptical confidence sets, linear contrasts, bootstrap score test, model misspecification.
    [Show full text]
  • On the Validity of the Formal Edgeworth Expansion for Posterior Densities
    Submitted to the Annals of Statistics ON THE VALIDITY OF THE FORMAL EDGEWORTH EXPANSION FOR POSTERIOR DENSITIES By John E. Kolassaz and Todd A. Kuffnerx Rutgers Universityz and Washington University in St. Louisx We consider a fundamental open problem in parametric Bayesian theory, namely the validity of the formal Edgeworth expansion of the posterior density. While the study of valid asymptotic expansions for posterior distributions constitutes a rich literature, the validity of the formal Edgeworth expansion has not been rigorously estab- lished. Several authors have claimed connections of various posterior expansions with the classical Edgeworth expansion, or have simply assumed its validity. Our main result settles this open problem. We also prove a lemma concerning the order of posterior cumulants which is of independent interest in Bayesian parametric theory. The most relevant literature is synthesized and compared to the newly-derived Edgeworth expansions. Numerical investigations illustrate that our expansion has the behavior expected of an Edgeworth expansion, and that it has better performance than the other existing expansion which was previously claimed to be of Edgeworth-type. 1. Introduction. The Edgeworth series expansion of a density function is a fundamental tool in classical asymptotic theory for parametric inference. Such expansions are natural refinements to first-order asymptotic Gaussian approximations to large-sample distributions of suitably centered and nor- malized functionals of sequences of random variables, X1;:::;Xn. Here, n is the available sample size, asymptotic means n ! 1, and first-order means that the approximation using only the standard Gaussian distribution incurs an absolute approximation error of order O(n−1=2).
    [Show full text]
  • Multivariate Generalized Gram-Charlier Series in Vector
    arXiv: math.ST: 1503.03212 Multivariate Generalized Gram-Charlier Series in Vector Notations Dharmani Bhaveshkumar C. Dhirubhai Ambani Institute of Information & Communication Technology (DAIICT), Gandhinagar, Gujarat, INDIA - 382001 e-mail: [email protected] Abstract: The article derives multivariate Generalized Gram-Charlier (GGC) series that expands an unknown joint probability density function (pdf ) of a random vector in terms of the differentiations of the joint pdf of a reference random vector. Conventionally, the higher order differentiations of a multivariate pdf in GGC series will require multi-element array or tensor representations. But, the current article derives the GGC series in vector notations. The required higher order differentiations of a multivari- ate pdf in vector notations are achieved through application of a specific Kronecker product based differentiation operator. Overall, the article uses only elementary calculus of several variables; instead Tensor calculus; to achieve the extension of an existing specific derivation for GGC series in uni- variate to multivariate. The derived multivariate GGC expression is more elementary as using vector notations compare to the coordinatewise ten- sor notations and more comprehensive as apparently more nearer to its counterpart for univariate. The same advantages are shared by the other expressions obtained in the article; such as the mutual relations between cumulants and moments of a random vector, integral form of a multivariate pdf, integral form of the multivariate Hermite polynomials, the multivariate Gram-Charlier A (GCA) series and others. Primary 62E17, 62H10; Secondary 60E10. Keywords and phrases: Multivariate Generalized Gram-Charlier (GGC) series; Multivariate Gram-Charlier A (GCA) series; Multivariate Vector Hermite Polynomials; Kronecker Product; Vector moments; Vector cumu- lants.
    [Show full text]
  • Sharp Composition Bounds for Gaussian Differential Privacy Via
    Sharp Composition Bounds for Gaussian Differential Privacy via Edgeworth Expansion Qinqing Zheng∗ Jinshuo Dong† Qi Long‡ Weijie J. Su§ University of Pennsylvania February 27, 2020 Abstract Datasets containing sensitive information are often sequentially analyzed by many algorithms. This raises a fundamental question in differential privacy regarding how the overall privacy bound degrades under composition. To address this question, we introduce a family of analytical and sharp privacy bounds under composition using the Edgeworth expansion in the framework of the recently proposed f-differential privacy. In contrast to the existing composition theorems using the central limit theorem, our new privacy bounds under composition gain improved tightness by leveraging the refined approximation accuracy of the Edgeworth expansion. Our approach is easy to implement and computationally efficient for any number of compositions. The superiority of these new bounds is confirmed by an asymptotic error analysis and an application to quantifying the overall privacy guarantees of noisy stochastic gradient descent used in training private deep neural networks. 1 Introduction Machine learning, data mining, and statistical analysis are widely applied to various applications impacting our daily lives. While we celebrate the benefits brought by these applications, to an alarming degree, the algorithms are accessing datasets containing sensitive information such as individual behaviors on the web and health records. By simply tweaking the datasets and leveraging the output of algorithms, it is possible for an adversary to learn information about and even identify certain individuals [FJR15, SSSS17]. In particular, privacy concerns become even more acute when the same dataset is probed by a sequence of algorithms.
    [Show full text]
  • Edgeworth Expansion and Saddle Point Approximation for Discrete Data with Application to Chance Games
    Degree Project Edgeworth Expansion and Saddle Point Approximation for Discrete Data with Application to Chance Games. Rani Basna 2010-09-27 Subject: Mathematics Level: Master Course code: 5MA11E Edgeworth Expansion and Saddle Point Approximation for Discrete Data With Application to Chance Games Abstract We investigate mathematical tools, Edgeworth series expansion and the saddle point method, which are approximation techniques that help us to estimate the distribution function for the standardized mean of independent identical distributed random variables where we will take into consideration the lattice case. Later on we will describe one important application for these mathematical tools where game developing companies can use them to reduce the amount of time needed to satisfy their standard requests before they approve any game. Keywords Characteristic function, Edgeworth expansion, Lattice random variables, Saddle point approximation. Acknowledgments First I would like to show my gratitude to my supervisor Dr.Roger Pettersson for his continuous support and help provided to me. It is a pleasure to me to thank Boss Media® for giving me this problem. I would like also to thank my fiancée Hiba Nassar for her encourage .Finally I want to thank my parents for their love and I dedicate this thesis to my father who inspired me. Rani Basna Number of Pages: 45 Contents 1 Introduction 3 2 Notations and Definitions 3 2.1 CharacteristicFunction. 3 2.2 CentralLimitTheorem. 6 2.3 Definition of Lattice Distribution and Bounded Variation . 7 3 Edgeworth Expansion 8 3.1 FirstCase ............................. 9 3.2 SecondCase............................ 14 3.2.1 AuxiliaryTheorems. 14 3.2.2 OnTheRemainderTerm.
    [Show full text]
  • Edgeworth Expansion and Bootstrap Approximation for M-Estimators of Linear Regression Parameters with Increasing Dimensions Muhammad Arif Tiro Iowa State University
    Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations 1991 Edgeworth expansion and bootstrap approximation for M-estimators of linear regression parameters with increasing dimensions Muhammad Arif Tiro Iowa State University Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Statistics and Probability Commons Recommended Citation Tiro, Muhammad Arif, "Edgeworth expansion and bootstrap approximation for M-estimators of linear regression parameters with increasing dimensions " (1991). Retrospective Theses and Dissertations. 10077. https://lib.dr.iastate.edu/rtd/10077 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]
  • UC Berkeley UC Berkeley Electronic Theses and Dissertations
    UC Berkeley UC Berkeley Electronic Theses and Dissertations Title Small Sample Inference Permalink https://escholarship.org/uc/item/4mh4741g Author Gerlovina, Inna Publication Date 2016 Supplemental Material https://escholarship.org/uc/item/4mh4741g#supplemental Peer reviewed|Thesis/dissertation eScholarship.org Powered by the California Digital Library University of California Small Sample Inference by Inna Gerlovina A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Biostatistics in the Graduate Division of the University of California, Berkeley Committee in charge: Professor Alan E. Hubbard, Chair Professor Mark J. van der Laan Professor Martyn T. Smith Fall 2016 Small Sample Inference Copyright 2016 by Inna Gerlovina 1 Abstract Small Sample Inference by Inna Gerlovina Doctor of Philosophy in Biostatistics University of California, Berkeley Professor Alan E. Hubbard, Chair Multiple comparisons and small sample size, common characteristics of many types of “Big Data” including those that are produced by genomic studies, present specific challenges that affect reliability of inference. Use of multiple testing procedures necessitates estimation of very small tail probabilities and thus approximation of distal tails of a test statistic distri- bution. Results based on large deviation theory provide a formal condition that is necessary to guarantee error rate control given practical sample sizes, linking the number of tests and the sample size; this condition, however, is rarely satisfied. Using methods that are based on Edgeworth expansions (relying especially on the work of Peter Hall), we explore what it might translate into in terms of actual error rates. Our investigation illustrates how far the actual error rates can be from the declared nominal levels, indicating poor error rate control.
    [Show full text]
  • A Uniform-In-$ P $ Edgeworth Expansion Under Weak Cram\'{E} R
    A UNIFORM-IN-P EDGEWORTH EXPANSION UNDER WEAK CRAM´ER CONDITIONS Kyungchul Song Vancouver School of Economics, University of British Columbia ABSTRACT. This paper provides a finite sample bound for the error term in the Edgeworth expansion for a sum of independent, potentially discrete, nonlattice random vectors, using a uniform-in-P version of the weaker Cram´er condition in Angst and Poly (2017). This finite sample bound can be used to derive an Edgeworth expansion that is uniform over the distributions of the random vectors. Using this result, we derive a uniform-in-P higher order expansion of resampling-based distributions. KEY WORDS. Edgeworth Expansion; Normal Approximation; Bootstrap; Resampling; Weak Cram´er Condition AMS 1991 CLASSIFICATION: 60E05, 62G20, 62E20 1. Introduction arXiv:1806.01431v2 [math.ST] 13 Aug 2019 Suppose that X n is a triangular array of independent random vectors taking values { i,n}i=1 Rd P n in and having mean zero. Let n be the collection of the joint distributions for (Xi,n)i=1. (One can view each joint distribution P in Pn as being induced by a probability measure n on a measurable space on which (Xi,n)i=1 live.) Define 1 n V Var (X ), n,P ≡ n P i,n i=1 X Date: August 14, 2019. I thank three anonymous referees for valuable comments in a previous version of this paper. All errors are mine. I acknowledge that this research was supported by Social Sciences and Humanities Research Council of Canada. Corresponding address: Kyungchul Song, Vancouver School of Economics, University of British Columbia, Vancouver, BC, Canada.
    [Show full text]
  • Expansions for Nearly Gaussian Distributions
    ASTRONOMY & ASTROPHYSICS MAY II 1998, PAGE 193 SUPPLEMENT SERIES Astron. Astrophys. Suppl. Ser. 130, 193–205 (1998) Expansions for nearly Gaussian distributions S. Blinnikov1,2 and R. Moessner2 1 Institute for Theoretical and Experimental Physics, 117259, Moscow, Russia Sternberg Astronomical Institute, 119899 Moscow, Russia 2 Max-Planck-Institut f¨ur Astrophysik, D-85740 Garching, Germany Received July 8; accepted November 10, 1997 Abstract. Various types of expansions in series of Cheby- Bertschinger 1991; Kofman et al. 1994; Moessner et al. shev-Hermite polynomials currently used in astrophysics 1994; Juszkiewicz et al. 1995; Bernardeau & Kofman 1995; for weakly non-normal distributions are compared, namely Amendola 1994; Colombi 1994; Moessner 1995; Ferreira the Gram-Charlier, Gauss-Hermite and Edgeworth expan- et al. 1997; Gazta˜naga et al. 1997), in analyzing the ve- sions. It is shown that the Gram-Charlier series is most locity distributions and fine structure in elliptical galaxies suspect because of its poor convergence properties. The (Rix & White 1992; van der Marel & Franx 1993; Gerhard Gauss-Hermite expansion is better but it has no intrinsic 1993; Heyl et al. 1994), and in studies of large Reynolds measure of accuracy. The best results are achieved with number turbulence (Tabeling et al. 1996). In this paper we the asymptotic Edgeworth expansion. We draw attention present a unified approach to the formalism used in those to the form of this expansion found by Petrov for arbitrary applications, illustrating it by examples of similar prob- order of the asymptotic parameter and present a simple al- lems arising in cosmology and in the theory of supernova gorithm realizing Petrov’s prescription for the Edgeworth line spectra.
    [Show full text]