Compound Poisson Approximation via Information Functionals

  title={Compound Poisson Approximation via Information Functionals},
  author={Andrew D. Barbour and Oliver Johnson and Ioannis Kontoyiannis and Mokshay M. Madiman},
An information-theoretic development is given for the problem of compound Poisson approximation, which parallels earlier treatments for Gaussian and Poisson approximation. Nonasymptotic bounds are derived for the distance between the distribution of a sum of independent integer-valued random variables and an appropriately chosen compound Poisson law. In the case where all summands have the same conditional distribution given that they are non-zero, a bound on the relative entropy distance… 

Figures and Tables from this paper

On the entropy of sums of Bernoulli random variables via the Chen-Stein method

  • I. Sason
  • Computer Science, Mathematics
    2012 IEEE Information Theory Workshop
  • 2012
Upper bounds on the error that follows from an approximation of this entropy by the entropy of a Poisson random variable with the same mean are derived and combines elements of information theory with the Chen-Stein method for Poisson approximation.

A criterion for the compound poisson distribution to be maximum entropy

It is shown that the compound Poisson does indeed have a natural maximum entropy characterization when the distributions under consideration are log-concave, which complements the recent development by the same authors of an information-theoretic foundation for compoundPoisson approximation inequalities and limit theorems.

Information in probability: Another information-theoretic proof of a finite de Finetti theorem

An upper bound on the relative entropy is derived between the distribution of the distribution in a sequence of exchangeable random variables, and an appropriate mixture over product distributions, using de Finetti’s classical representation theorem as a corollary.

Entropy bounds for discrete random variables via coupling

  • I. Sason
  • Mathematics, Computer Science
    2013 IEEE International Symposium on Information Theory
  • 2013
New bounds on the difference between the entropies of two discrete random variables in terms of the local and total variation distances between their probability mass functions are provided.

Log-concavity, ultra-log-concavity, and a maximum entropy property of discrete compound Poisson measures

Mutual Information, Relative Entropy, and Estimation in the Poisson Channel

Relative entropy quantifies the excess estimation loss due to mismatch in this setting, parallel to those recently found for the Gaussian channel: the I-MMSE relationship of Guo, the relative entropy and mismatched estimation relationship of Verdú, and the relationship between causal and noncasual mismatch estimation of Weissman.

Improved lower bounds on the total variation distance and relative entropy for the Poisson approximation

  • I. Sason
  • Mathematics, Computer Science
    2013 Information Theory and Applications Workshop (ITA)
  • 2013
New lower bounds on the total variation distance between the distribution of a sum of independent Bernoulli random variables and the Poisson random variable (with the same mean) are derived via the

Compound Poisson Approximation to Convolutions of Compound Negative Binomial Variables

In this paper, the problem of compound Poisson approximation to the convolution of compound negative binomial distributions, under total variation distance, is considered. First, we obtain an error

Accurate Inference for the Mean of the Poisson-Exponential Distribution

Although the random sum distribution has been well-studied in probability theory, inference for the mean of such distribution is very limited in the literature. In this paper, two approaches are

Entropy and the fourth moment phenomenon



On the entropy and log-concavity of compound Poisson measures

It is shown that the natural analog of the Poisson maximum entropy property remains valid if the measures under consideration are log-concave, but that it fails in general.

Compound Poisson Approximation for Nonnegative Random Variables Via Stein's Method

The aim of this paper is to extend Stein's method to a compound Poisson distribution setting. The compound Poisson distributions of concern here are those of the form POIS$(\nu)$, where $\nu$ is a

An expansion in the exponent for compound binomial approximations

The purpose of this paper is two-fold. First, we introduce a new asymptotic expansion in the exponent for the compound binomial approximation of the generalized Poisson binomial distribution. The

Compound Poisson approximation: a user's guide

Compound Poisson approximation is a useful tool in a variety of applications, including insurance mathematics, reliability theory, and molecu- lar sequence analysis. In this paper, we review the ways

Measure Concentration for Compound Poisson Distributions

We give a simple development of the concentration properties of compound Poisson measures on the nonnegative integers. A new modification of the Herbst argument is applied to an appropriate modified

Accuracy of approximation in the Poisson theorem in terms of the χ2-distance

We study the limit behavior of the χ2-distance between the distributions of the nth partial sum of independent not necessarily identically distributed Bernoulli random variables and the accompanying

Generalized Entropy Power Inequalities and Monotonicity Properties of Information

A simple proof of the monotonicity of information in central limit theorems is obtained, both in theSetting of independent and identically distributed (i.i.d.) summands as well as in the more general setting of independent summands with variance-standardized sums.

Stein's Method: Expository Lectures and Applications

A review of Stein’s method applied to the case of discrete random variables and attempt to complete one of Stein's open problems, that of providing a discrete version for chapter 6 of his book.

On the Entropy of Compound Distributions on Nonnegative Integers

  • Yaming Yu
  • Mathematics
    IEEE Transactions on Information Theory
  • 2009
Two recent results of Johnson (2008) on maximum entropy characterizations of compound Poisson and compound binomial distributions are proved under fewer assumptions and with simpler arguments.

On the Distribution of Sums of Independent Random Variables

Let {X j ; j = 1, 2,...} be a finite sequence of independent random variables. Let S = Σ X j be their sum, and let P j be the distribution of X j . Let M be the measure defined on the line deprived