scientific article; zbMATH DE number 3980131
In 1953 A. N. Kolmogorov formulated the problem of approximating the convolutions of distribution functions \(*^{n}_{i=1}F_ i\) by infinitely divisible distributions. Many mathematicians payed their attention to this problem. It got its final solution by common efforts of A. N. Kolmogorov himself, Yu. V. Prokhorov, F. M. Kagan, I. P. Tsaregradsky, L. D. Meshalkin, Yu. P. Studnev, I. A. Ibragimov, E. L. Presman and the present authors. The final solution of the problem under discussion in the case of identical distribution functions \(F_1 = F_2= \ldots = F_n\) was obtained by the first author with the help of his ``triangular functions method. The two following theorems play the main role in the book under review. They present the modern solution of Kolmogorov's problem: Theorem 1. a) For any positive integer \(n\) and distribution function \(F\) there exists an infinitely divisible distribution function \(D_n\) such that \[ \rho (F^{*n},D_n)=\max_{x}| F^{*n}(x)-D_n(x)| \leq C_1 n^{-2/3}. \] b) For any positive integer \(n\) there exists a distribution function \(F_n\) such that \[ \inf_{D\in {\mathcal D}}\rho (F_{n}^{*n},D)\geq C_2 n^{-2/3} \] where \(\mathcal D\) is the class of all infinitely divisible distribution functions and \(C_1\), \(C_2\) are positive constants. Denote Levy's metric by \(L\) and the distribution function with the unit jump in the point a by \(E_a\). Theorem 2. a) Let distribution functions \(F_i\), \(i=1,2,\ldots,n\), satisfy the inequalities \(L(F_i,E_{\beta_i})\leq \varepsilon\) for some \(\beta_i\), \(i=1,2,\ldots,n\), and \(\varepsilon >0\). Then there exists an infinitely divisible distribution function \(D\) such that \[ L(*^{n}_{i=1}F_i,D)\leq C_3\varepsilon (| \ln \varepsilon | +1) \tag{1} \] b) For any \(\varepsilon\in (0,1)\) there exist a distribution function \(F\) and positive integer \(n\) such that \(L(F,E_ 0)\leq \varepsilon\) and \[ \inf_{D\in {\mathcal D}}L(F^{*n},D)\geq C_4\varepsilon (| \ln \varepsilon | +1). \] Here \(C_3\) and \(C_4\) are positive constants. Note that (1) turns out to be valid if \(D\) is an accompanying infinitely divisible distribution for \(*^{n}_{i=1}F_i\). The main achievement that made it possible to solve Kolmogorov's problem is the proof of some new inequalities for concentration functions of sums of independent random variables. The distinctive feature of these inequalities is that they take into account the structure of distribution functions and, in particular, the structure of their supports. The monograph is divided into eight chapters. Chapter 1 contains preliminary information. Chapter 2 deals with the necessary properties of concentration functions including the authors' results mentioned above. Methods of estimating the uniform metric are discussed in chapter 3. Here the ``triangular function method is described. Chapter 4 contains some results concerning the infinitely divisible approximations to binomial distributions. The estimates of the uniform metric and Levy's metric in the set of distribution functions are presented in chapters 5 and 6. Chapters 7 and 8 contain the proofs of the main results. The book under review presents various stuff concerning the sums of independent random variables which is applied to the proof of the main results. Some known results are reproved with the help of new more easily understood methods. Readers can find many interesting problems for their own investigation. Bibliographic comments help the reader to guide himself through the history of Kolmogorov's problem and some neighbouring questions.
- On asymptotic proximity of distributions
- Estimates for the closeness of successive convolutions of multidimensional symmetric distributions
- Approximation of mixture of distributions
- Certain class of nonuniform estimates in multidimensional limit theorems
- Bounds for approximations of \(n\)-fold convolutions of distributions with unlimited divisibility and the moments problem
- Arak's inequalities for the generalized arithmetic progressions
- An estimate of the absolute constant in the inequality for the uniform distance between the distributions of sequential sums of independent random variables
- Why intervals? A simple limit theorem that is similar to limit theorems from statistics
- On asymptotic expansions in the first uniform Kolmogorov theorem
- On multivariate Le Cam theorem and compound Poisson measures
- Quantile coupling inequalities and their applications
- Poisson approximation in terms of the Gini-Kantorovich distance
- Compound Poisson approximation
- On the \(T\)-test
- Rare events and Poisson point processes
- On improvement of the estimate for the distance between sequential sums of independent random variables
- An improved multivariate version of Kolmogorov's second uniform limit theorem
- On calculation of constants in the Arak inequalities for concentration functions of convolutions of probability distributions
- On approximation of convolutions by accompanying laws in the scheme of series
- On the accuracy of Poisson approximation
- Khinchine's theorem and Edgeworth approximations for weighted sums
- Uniform approximation of the integrated density of states for long-range percolation Hamiltonians
- Approximation of a sample by a Poisson point process
- Berry-Esseen bounds for standardized subordinators via moduli of smoothness
- Calculation of constants in lemma on the functions \(W(X)\) and \(G(T)\) in the method of smooth triangular functions
- Free infinitely divisible approximations of \(n\)-fold free convolutions
- The uniform convergence of the sum of independent random variables
- Topics in probability
- On a problem of A. N. Kolmogorov
- Limit theorems for the sums of independent random variables.
- scientific article; zbMATH DE number 4007337 (Why is no real title available?)
- scientific article; zbMATH DE number 4050639 (Why is no real title available?)
- scientific article; zbMATH DE number 4078401 (Why is no real title available?)
- scientific article; zbMATH DE number 4078402 (Why is no real title available?)
- scientific article; zbMATH DE number 4088618 (Why is no real title available?)
- scientific article; zbMATH DE number 4109736 (Why is no real title available?)
- scientific article; zbMATH DE number 195083 (Why is no real title available?)
- On alternative approximating distributions in the multivariate version of Kolmogorov's second uniform limit theorem
- Refined total variation bounds in the multivariate and compound Poisson approximation
- An optimal Berry-Esseen type theorem for integrals of smooth functions
- Convergence to infinite-dimensional compound Poisson distributions on convex polyhedra
- Calculation of Constants in Lemma on Pseudometric for One-Dimensional Method of Smooth Triangular Functions
- On Poisson approximation
- Smoothness and Lévy concentration function inequalities for distributions of random diagonal sums
- Asymptotics in multivariate le Cam's theorem
- Multivariate Hipp-type compound Poisson approximations for lattice distributions. I
- Approximating compound Poisson binomial distribution under 1-dependence
- Methods of Poisson approximation
- Multivariate estimates for the concentration functions of weighted sums of independent, identically distributed random variables
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3744970)