Estimates of the proximity of a distribution to the normal law

Institute of Mathematics and Cybernetics, Academy of Sciences of the Lithuanian SSR. Vilnius University, Vilnius Civil Engineering Institute. Translated from Litovskii Matematicheskii Sbornik (Lietuvos Matematikos Rinkinys), Vol. 29, No. 4, pp. 657–673, October–December, 1989.     

On the normal approximation of a binomial random sum

We present upper bounds of the L _s norms of the normal approximation for random sums of independent identically distributed random variables X ₁ , X ₂ , . . . with finite absolute moments of order 2 + δ, 0 < δ ≤ 1, where the number of summands N is a binomial random variable independent of the summands X ₁ , X ₂ , . . . . The upper bounds obtained are of order (E N) ^?δ/2 for all 1 ≤ s ≤ ∞.     

Test of misspecification with application to negative binomial distribution

A misspecification test based directly on Bartlett’s First Identity is examined. This test is exemplified by the negative binomial distribution. A Monte Carlo simulation study has been conducted, in the context of testing distributional misspecification, and the performance of the proposed test has been benchmarked with some goodness-of-fit tests based on the empirical distribution function. The results suggest that the proposed test is viable in terms of computational speed and statistical power, and has the advantage that complications arising from the use of the covariance matrix in White’s information matrix test are avoided.     

Limit theorems for the multivariate binomial distribution

Nonsingular limit distributions are determined for sequences of affine transformations of random vectors whose distributions are multivariate binomial. Each of these limit distributions is that of an affine transformation of a random vector having a multivariate normal distribution or a multivariate Possion distribution or a joint distribution of two independent random vectors, one normal and the other Poisson.     

Efficient computation for the Poisson binomial distribution

Direct construction of the probability distribution function for a Poisson binomial random variable, where success probabilities may vary from trial to trial, requires on the order of \(2^{n}\) calculations, and is computationally infeasible for all but modest sized problems. An approach offered by Thomas and Taub (J Stat Comput Simul 14:125–131 1982) reduces this effort to approximately \(n^{2}\) , which, while certainly an improvement, can still be significant for large values of \(n\) . We offer modifications to the method of Thomas and Taub that greatly reduce the computations while still delivering highly accurate results.     

On the binomial arithmetical rank

The binomial arithmetical rank of a binomial ideal I is the smallest integer s for which there exist binomials f₁,..., f_s in I such that rad (I) = rad (f₁,..., f_s). We completely determine the binomial arithmetical rank for the ideals of monomial curves in P_KⁿP_K^n. In particular we prove that, if the characteristic of the field K is zero, then bar (I(C)) = n - 1 if C is complete intersection, otherwise bar (I(C)) = n. While it is known that if the characteristic of the field K is positive, then bar (I(C)) = n - 1 always.     

On one stochastic model that leads to a stable distribution

We consider an integral equation describing the contagion phenomenon, in particular, the equation of the state of a hereditarily elastic body, and interpret this equation as a stochastic model in which the Rabotnov exponent of fractional order plays the role of density of probability of random delay time. We invesgigate the approximation of the distribution for sums of values with a given density to the stable distribution law and establish the principal characteristics of the corresponding renewal process. Ukrainian Mining Academy, Dnepropetrovsk. Translated from Ukrainskii Matematicheskii Zhurnal, Vol. 49, No. 11, pp. 1572–1579, November, 1997.     

On congruences related to central binomial coefficients

It is known that

     

On the general excess bound for binary codes with covering radius one

Let K(n,1)$K(n,1)$ denote the minimal cardinality of a binary code of length n$n$ and covering radius one. Fundamental for the theory of lower bounds for K(n,1)$K(n,1)$ is the covering excess method introduced by Johnson and van Wee. Let _δi${\delta }_i$ denote the covering excess on a sphere of radius i$i$, 0≤i≤n$0\le i\le n$. Generalizing an earlier result of van Wee, Habsieger and Honkala showed _δp−1≥p−1${\delta }_{p-1}\ge p-1$ whenever n≡−1$n\equiv -1$ (mod p$p$) for an odd prime p$p$ and _δ0=_δ1=?=_δp−2=0${\delta }_0={\delta }_1=?={\delta }_{p-2}=0$ holds. In the present paper we give the new estimation _δp−1≥(p−2)p−1${\delta }_{p-1}\ge (p-2)p-1$ instead. This answers a question of Habsieger and yields a “general improvement of the general excess bound” for binary codes with covering radius one. The proof uses a classification theorem for certain subset systems as well as new congruence properties for the δ$\delta$-function, which were conjectured by Habsieger.     

Tight lower bound on the probability of a binomial exceeding its expectation

We give the proof of a tight lower bound on the probability that a binomial random variable exceeds its expected value. The inequality plays an important role in a variety of contexts, including the analysis of relative deviation bounds in learning theory and generalization bounds for unbounded loss functions.     

Poisson type approximations for the Markov binomial distribution

The Markov binomial distribution is approximated by the Poisson distribution with the same mean, by a translated Poisson distribution and by two-parametric Poisson type signed measures. Using an adaptation of Le Cam’s operator technique, estimates of accuracy are proved for the total variation, local and Wasserstein norms. In a special case, asymptotically sharp constants are obtained. For some auxiliary results, we used Stein’s method.     

Convergence of one partial form of sums of dependent random vectors to a normal distribution

The method of moments is used to establish an asymptotic normal distribution for a sequence of dependent random vectors of special form.Translated fromMatematicheskii Zametki, Vol. 5, No. 6, pp. 691–695, June, 1969.     

A compound of the generalized negative binomial distribution with the generalized beta distribution

This paper presents a compound of the generalized negative binomial distribution with the generalized beta distribution. In the introductory part of the paper, we provide a chronological overview of recent developments in the compounding of distributions, including the Polish results. Then, in addition to presenting the probability function of the compound generalized negative binomial-generalized beta distribution, we present special cases as well as factorial and crude moments of some compound distributions.