Moments of the generalized hyperbolic distribution

In this paper we demonstrate a recursive method for obtaining the moments of the generalized hyperbolic distribution. The method is readily programmable for numerical evaluation of moments. For low order moments we also give an alternative derivation of the moments of the generalized hyperbolic distribution. The expressions given for these moments may be used to obtain moments for special cases such as the hyperbolic and normal inverse Gaussian distributions. Moments for limiting cases such as the skew hyperbolic t and variance gamma distributions can be found using the same approach.     

Moments and cumulants of the multivariate normal distribution

Expressions in vector notation are given for the central moments, the non–central moments and the cumulants of arbitrary order of the multivariate normal distribution     

Moments of the support weight distribution of linear codes

In this work, we give the expectation and the covariance formulas for the support weight distributions of linear codes.     

Moments and distribution of the local time of a two-dimensional random walk

Let ?(n,x)$?(n,x)$ be the local time of a random walk on Z²${\mathbb{Z}}^2$. We prove a strong law of large numbers for the quantity _Ln(α)=_∑x∈Z2?(n,x)^α$L_n\left(\alpha \right)={\sum }_{x\in {\mathbb{Z}}^2}?{(n,x)}^{\alpha }$ for all α≥0$\alpha \ge 0$. We use this result to describe the distribution of the local time of a typical point in the range of the random walk.     

Moments of Change

There is a strong intuition that for a change to occur, there must be a moment at which the change is taking place. It will be demonstrated that there are no such moments of change, since no state the changing thing could be in at any moment would suffice to make that moment a moment of change. A moment in which the changing thing is simply in the state changed from or the state changed to cannot be the moment of change, since these states are respectively before and after the change; moreover, to select one of these moments over the other as the moment of change would be arbitrary. A moment in which the changing thing is neither in the state changed from nor in the state changed to cannot be the moment of change, since there are changes for which it is impossible for something to be in neither state. Finally, the moment of change cannot be a moment in which the changing thing is in both the state changed from and the state changed to, as suggested by Graham Priest and others. Even if, like proponents of this view, we are willing to accept the contradictions that the account entails, it is demonstrated that on such a model, every change would require an infinite number of other changes, every change would take an infinite amount of time, and some changes would occur without occurring at any time. Further, the model is grossly counterintuitive, with the exact nature of the counterintuitive element depending on what model of time and space one endorses. Finally, it is demonstrated that this model is incompatible with the Leibniz Continuity Condition.     

Integral Moments of L-Functions

We give a new heuristic for all of the main terms in the integralmoments of various families of primitive L-functions. The resultsagree with previous conjectures for the leading order terms.Our conjectures also have an almost identical form to exactexpressions for the corresponding moments of the characteristicpolynomials of either unitary, orthogonal, or symplectic matrices,where the moments are defined by the appropriate group averages.This lends support to the idea that arithmetical L-functionshave a spectral interpretation, and that their value distributionscan be modelled using Random Matrix Theory. Numerical examplesshow good agreement with our conjectures. 2000 Mathematics SubjectClassification 11M26, 15A52.     

Moments of square-integrable functions

This paper is motivated by [2], where we have given necessary and sufficient conditions for a given basis P in the space of polynomials to be orthogonal with respect to the measure ϱdφ for a certain function ϱ ϵ L₂(dφ). Let P = {p_i: i = 0, 1, …}, p₀ = 1. Then the conditions are (1) a multivariate analog of the three-term recurrence relation holds, see Section 4 for details; and (2) {q_i = ∑_{j = 0}^∞ c_ij P_j, i = 0, 1, …} is a φ-orthonormal basis in the space of polynomials for some coefficients c_ij such that ∑_{i = 0}^∞ c_i0² <-∞. This paper provides an algebraic condition (a condition on the coefficients c_i0) such that ϱ satisfies ∥p∥ <B, Bϵ (0, ∞], and has a cone-positivity property. In particular, our results imply that ϱ is nonnegative a.e. if ∑_{i = 0}^∞ c_i0² < ∞ and ∑_{jϵ Sk} c_j0 q_j defines nonnegative polynomials for certain finite sets S₁, S₂, … of integers.     

Improved approximations to distributions of the largest and the smallest latent roots of a wishart matrix

Summary Normalizing transformations of the largest and the smallest latent roots of a sample covariance matrix in a normal sample are obtained, when the corresponding population roots are simple. Using our results, confidence intervals for population roots may easily be constructed. Some numerical comparisons of the resulting approximations are made in a bivariate case, based on exact values of the probability integral of latent roots.     

Stochastic Generalized Method of Moments

The generalized method of moments (GMM) is a very popular estimation and inference procedure based on moment conditions. When likelihood-based methods are difficult to implement, one can often derive various moment conditions and construct the GMM objective function. However, minimization of the objective function in the GMM may be challenging, especially over a large parameter space. Due to the special structure of the GMM, we propose a new sampling-based algorithm, the stochastic GMM sampler, which replaces the multivariate minimization problem by a series of conditional sampling procedures. We develop the theoretical properties of the proposed iterative Monte Carlo method, and demonstrate its superior performance over other GMM estimation procedures in simulation studies. As an illustration, we apply the stochastic GMM sampler to a Medfly life longevity study. Supplemental materials for the article are available online.     

Moments of Markov Random Evolutions

We find moments of a process of Markov random evolutions in a finite-dimensional space.     

Moments of Generalized Wishart Distributions

Moments of generalized Wishart distributions are obtained through tensor differential forms and reindexing.