On the rate of convergence of the distributions of certain statistics to the Laplace and student distributions

This work deals with asymptotically normal statistics. If the size of sample is replaced by a random variable that is the maximum of n independent random variables with Pareto discrete distribution, then the distribution of these statistics is asymptotically Laplacian. If the size of the sample is replaced by a random variable with negative binomial distribution, then the distribution of these statistics is asymptotically a Student distribution. In the present study, the rates of convergences of these statistics’ distributions to the corresponding limiting distributions are estimated and the constants from expressions for estimating the rates are determined more accurately.     

On the rate of convergence of the distributions of certain statistics to the Laplace distribution

An estimation of the rate of convergence of the distributions of asymptotically normal statistics based on a random sample of random size to the Laplace distribution is obtained. It is assumed that the random size of a sample does not depend on the members of the sample and has a number of special properties.     

The rate of convergence for subexponential distributions

A distribution functionF on the nonnegative real line is called subexponential if

A note on the rate of convergence for certain nonlinear evolution equations

We estimate the rate of convergence to the stationary state for evolution equations which admit many H-theorems with respect to this stationary state.     

A rate of convergence in the poissonian representation of stable distributions

Published in Lietuvos Matematikos Rinkinys, Vol. 36, No. 4, pp. 486–500, October–December, 1996.     

Parameter estimation for convolutions and compound distributions

In this paper, we consider estimating the parameters of convolutions and compound distributions using their components. We show that the component estimators of variance are always at least as efficient as direct estimators, and investigate the effects of changes in the components upon their elative efficiency.     

On the implication of a certain rate of convergence to normality

Probability Theory and Related Fields -     

Minimax estimation of location parameters for certain spherically symmetric distributions

Families of minimax estimators are found for the location parameters of a p-variate distribution of the form

, where G(·) is a known c.d.f. on (0, ∞), p ≥ 3 and the loss is sum of squared errors. The estimators are of the form $\text{(1 ? ar(X′X)}\text{E}{}_0\text{(}\text{1}\text{X′X}\text{)X′X)X}$ where 0 ≤ a ≤ 2, r(X′X) is nondecreasing, and $\text{r(X′X)}\text{X′X}$ is nonincreasing. Generalized Bayes minimax estimators are found for certain G(·)'s.     

Approximation and estimation of some compound distributions

The distribution of the total amount claimed up to time t can often be written in the form of a compound distribution G_t(x) = Σp_n(t)F⁽ⁿ⁾(x) where p_n(t) is the probability of exactly n claims while F is the distribution of a single claim. In the actuarial literature one often finds approximations of G_t(x) when the time t is large. It seems more natural to take t fixed and to look for approximations for x large. This paper contains a number of such results for a Poisson process and for a Pascal process. Different hypotheses on the tail behaviour of F(t) yield different expressions to estimate 1 - G_t(x). The results obtained should prove to have wider applicability than suggested by the insurance context. Within it, however, applications to premium calculation principles are immediate.     

On the rate of convergence of certain methods of centers

It is shown in this paper that a theoretical method of centers, introduced by Huard, converges linearly. It is also shown, by counter-example, that a modified method of centers due to Huard and a method of feasible direction due to Topkis and Veinot cannot converge linearly even under convexity assumptions. Because of this, a new modified method of centers is introduced which uses a quadratic programming direction finding subroutine. In most uses this new method is not more complicated than Huard's modified method of centers. But it does converge linearly. A method for implementing it without loss of rate of convergence is also discussed.Research sponsored by the Joint Services Electronics Program, Grant AF-AFOSR-68-1488 and the National Aeronautics and Space Administration, Grant NGL-05-003-016.     

Nonparametric estimation of compound distributions with applications in insurance

A nonparametric estimator of the distribution functionG of a random sum of independent identically distributed random variables, with distribution functionF, is proposed in the case where the distribution of the number of summands is known and a random sample fromF is available. This estimator is found by evaluating the functional that mapsF ontoG at the empirical distribution function based on the random sample. Strong consistency and asymptotic normality of the resulting estimator in a suitable function space are established using appropriate continuity and differentiability results for the functional. Bootstrap confidence bands are also obtained. Applications to the aggregate claims distribution function and to the probability of ruin in the Poisson risk model are presented.     

The convergence rate for the strong law of large numbers: General lattice distributions

LetX ₁,X ₂, ... be a sequence of independent random variables with common lattice distribution functionF having zero mean, and let (S _n ) be the random walk of partial sums. The strong law of large numbers (SLLN) implies that for any and >0