Large Deviation Probabilities for Random Walks with Semiexponential Distributions

Siberian Mathematical Journal -     

Some Asymptotic Results for One-Sided Large Deviation Probabilities

We investigate the asymptotic behavior of the sum of independent real random variables. We assume that the random variables are not identically distributed but the average of distribution functions of these random variables is equivalent to some heavy-tailed limit distribution function. An example with Pareto law as limit function is given.     

Asymptotic Expansions for Large Deviation Probabilities of Noncentral Generalized Chi-Square Distributions

Asymptotic expansions for large deviation probabilities are used to approximate the cumulative distribution functions of noncentral generalized chi-square distributions, preferably in the far tails. The basic idea of how to deal with the tail probabilities consists in first rewriting these probabilities as large parameter values of the Laplace transform of a suitably defined function f_k; second making a series expansion of this function, and third applying a certain modification of Watson's lemma. The function f_k is deduced by applying a geometric representation formula for spherical measures to the multivariate domain of large deviations under consideration. At the so-called dominating point, the largest main curvature of the boundary of this domain tends to one as the large deviation parameter approaches infinity. Therefore, the dominating point degenerates asymptotically. For this reason the recent multivariate asymptotic expansion for large deviations in Breitung and Richter (1996, J. Multivariate Anal.58, 1–20) does not apply. Assuming a suitably parametrized expansion for the inverse g⁻¹ of the negative logarithm of the density-generating function, we derive a series expansion for the function f_k. Note that low-order coefficients from the expansion of g⁻¹ influence practically all coefficients in the expansion of the tail probabilities. As an application, classification probabilities when using the quadratic discriminant function are discussed.     

Large Deviation Probabilities for Square-Gaussian Stochastic Processes

We investigate properties of square-Gaussian stochastic processes. These processes are formed by quadratic forms of Gaussian processes or by limits in the mean square of quadratic forms of Gaussian processes. Special classes of these processes are determined and investigated. For processes from these classes estimates of large deviation probability are obtained. These estimates we use to estimate the probability that Gaussian vector-valued process leave some region on some interval of time. We construct asymptotic confidence regions for estimates of covariance functions of vector-valued Gaussian processes. Criterion of hypothesis testing on covariance functions of these processes is constructed.     

A Lower Bound of Large-Deviation Probabilities for the Sample Mean under the Cramer Condition

Let X₁, X₂, ... be i.i.d. random variables satisfying the condition

一类随机场越界概率的大偏差

For Y₁, Y₂,…i .i .d . with Y₁～N(μ, 1) and S_n=(?)Y_i, the large deviations are obtained for theprobabilities that(?) conditionally given (i) S_m=0, and (ii) S_mi=ξ. Applied these results to the double change points model with some nuisance parameters, we developed the large deviation for the significance level of the likelihood ratio test.     

多项分布与Dirichlet分布概率的相互表示

讨论了四种多项分布尾概率与四种Dirichlet分布尾概率的相互表示,并将结果应用于Majorization理论,得到了多项分布和Dirichlet分布对应的一些性质.同时,结果可应用于多项分布最大、最小参数的贝叶斯推断,在佛罗里达州沃尔顿县白人和黑人的职业状态调查结果中,求出最大、最小参数的后验分布函数以及95%贝叶斯区间估计,模拟结果表明提供的方法具有较好的表现.     

Semiexponential Distributions and Related Large Deviation Principles for Trajectories of Random Walks

Siberian Mathematical Journal - We obtain a rather simple characterization of semiexponential distributions. This allows us to relax substantially the conditions for the fulfillment of the...     

Some Classes of Life Distributions Characterized by Concave Ordering

1.IntroductionandDefintionsInreliabilitytheory,maintenancetheory,biometricsandeconometrics,variousclassesoflifedistributionsbasedondifferentconceptsofagingareveryuseful.Recently,anumberofclassesoflifedistributionshavebeenintroducedbymeansofdifferentapproachestocharacterizingagingproperty.Oneoftheseapproachesisbasedonthefactthatmanyclassesoflifedistributionmaybecharacterizedbyvariousstochasticorderings,see,e.g.[if.[2],[3]and[4],etc.LetFbethedistributionfunctionofanon-negativerandomvariableXre…     

Some Classes of Multivariate Life Distributions in Discrete Time

New classes of multivariate survival distribution functions based on monotonic behaviour of a multivariate failure rate are developed in the discrete set up. Relationship among the classes along with multivariate geometric distributions that act as boundaries of the various classes are identified.     

Logarithmic Level Comparison for Small Deviation Probabilities

Log-level comparisons of the small deviation probabilities are studied in three different but related settings: Gaussian processes under the L ² norm, multiple sums motivated by tensor product of Gaussian processes, and various integrated fractional Brownian motions under the sup-norm. Wenbo V. Li: Supported in part by NSF Grant DMS-0505805.     

Logarithmic Level Comparison for Small Deviation Probabilities

Log-level comparisons of the small deviation probabilities are studied in three different but related settings: Gaussian processes under the L² norm, multiple sums motivated by tensor product of Gaussian processes, and various integrated fractional Brownian motions under the sup-norm. An erratum to this article can be found at     

Inferences on Correlation Coefficients in Some Classes of Nonnormal Distributions

Correlation coefficients have many applications for studying the relationship among multivariate observations. Classical inferences on correlation coefficients are mainly based on the normality assumption. This assumption is hardly realistic in the real world, which implies that the procedures on correlation coefficients used in many statistical software packages may not be relevant to most data sets in practice. However, we show that the classical procedures, possibly after simple corrections, are also valid in classes of distributions with large skewnesses and heterogeneous marginal kurtoses. A useful class of nonnormal distributions is identified for each of several types of correlation coefficients. The marginals of these distributions may include a variety of univariate distributions with different shapes. The results generalize the classical procedures to much larger classes of distributions than previously known and give a better understanding of the historical controversy regarding the behavior of the sample correlation coefficient. An implication is that one need not be worried so much by the nonnormality of data sets when using these classical procedures, providing simple corrections are evaluated and possibly undertaken.     

On Structural and Asymptotic Properties of Some Classes of Distributions

We consider the two-parametric classes of distributions which comprise exponential dispersion models possessing a simple structure. Some of our models are related to Galton-Watson processes and branching diffusions. We study asymptotic properties of related classes of distributions emphasizing their convergence to Poisson-exponential law. We establish/refute infinite divisibility for certain classes of distributions. The critical value for the deflation of zeros from a geometric law which preserves infinite divisibility is determined. A class of distributions related to zero-modified geometric and logarithmic laws is considered.     

Large Deviation Principle for Exchangeable Sequences: Necessary and Sufficient Condition

For an exchangeable sequence of random variables valued in a Polish space, we obtain a necessary and sufficient condition for the large deviation principles of the occupation measure L _n :=(1/n) and of the process-level empirical measures.     

广义复合二项风险模型的若干大偏差结果

This paper is a further investigation of large deviation for partial and random sums of random variables, where {Xn,n ≥ 1} is non-negative independent identically distributed random variables with a common heavy-tailed distribution function F on the real line R and finite mean μ∈ R. {N（n）,n ≥ 0} is a binomial process with a parameter p ∈ （0,1） and independent of {Xn,n ≥ 1}; {M（n）,n ≥ 0} is a Poisson process with intensity λ 〉 0, Sn = ΣNn i=1 Xi-cM（n）. Suppose F ∈ C, we futher extend and improve some large deviation results. These results can apply to certain problems in insurance and finance.     

Sensitivity of the Stability Bound for Ruin Probabilities to Claim Distributions

Methodology and Computing in Applied Probability - We are interested in the approximation of the ruin probability of a classical risk model using the strong stability method. Particularly, we study...     

An Inequality for Large Deviation Probabilities of Sums of Bounded i.i.d. Random Variables

We provide precise bounds for tail probabilities, say {M _n x}, of sums M _n of bounded i.i.d. random variables. The bounds are expressed through tail probabilities of sums of i.i.d. Bernoulli random variables. In other words, we show that the tails are sub-Bernoullian. Sub-Bernoullian tails are dominated by Gaussian tails. Possible extensions of the methods are discussed.