Simultaneous confidence bands for expectile functions

Expectile regression, as a general M smoother, is used to capture the tail behaviour of a distribution. Let (X ₁,Y ₁),…,(X _n ,Y _n ) be i.i.d. rvs. Denote by v(x) the unknown τ-expectile regression curve of Y conditional on X, and by v _n (x) its kernel smoothing estimator. In this paper, we prove the strong uniform consistency rate of v _n (x) under general conditions. Moreover, using strong approximations of the empirical process and extreme value theory, we consider the asymptotic maximal deviation sup_0≤x≤1|v _n (x)?v(x)|. According to the asymptotic theory, we construct simultaneous confidence bands around the estimated expectile function. Furthermore, we apply this confidence band to temperature analysis. Taking Berlin and Taipei as an example, we investigate the temperature risk drivers to these two cities.     

Strict stationarity of ar(p) processes generated by nonlinear random functions with additive perturbations

Let {X_n} be a generalized autoregressive process of order ρ defined by X_n=Φ_n(X_n-ρ,…,X_n-1)-η_m, where {φ_n} is a sequence of i.i.d. random maps taking values on H, and {η_n} is a sequence of i.i.d. random variables. Let H be a collection of Borel measurable functions on R_P to R. By considering the associated Markov process, we obtain sufficient conditions for stationarity, (geometric) ergodicity of {X_n}.     

Revealing the dependence structure between X(1) and X(n)

In this paper we address the dependence structure of the minimum and maximum of n iid random variables X₁,…,X_n by determining their copula. It is then easy to give an alternative proof for their asymptotic independence and to calculate Kendall's τ and Spearman's ρ for (X₍₁₎,X_(n)). This will show that the dependence between the variables is already small for small sample sizes. Finally, it can be shown that 3τ_n⩾ρ_n⩾τ_n>0. Although closed-form expressions are available for τ_n and ρ_n, we cannot compare them directly but have to use the concept of positive likelihood ratio dependence to establish this result.     

Stochastic orderings related to ranking in an nth order record process

Let X₁, X₂, …, X_m be successive observations on m objects, numbered 1,2, …, m. If X₁ belongs to the n largest observations among X₁,X₂,…,X_i, object i is called a record, i = 1,2,…,m; n?1.In investigating the influence of the ranking of the objects on the expected number of records, a hierarchy of stochastic order relations between random variables arises.It is these order relations and their relationship with known stochastic orderings that are studied in this paper.     

A unified approach to characterization problems using conditional expectations

In recent years characterization problems have become of increasing interest. It is well known that mean residual life e(x) = E(X - x|X ⩾x) and right-censored mean function m_R(x) = E(X | X ⩽ x), uniquely determine the distribution function F(x) = P(X ⩽ x). In this paper, we study characterizations problems for general distributions, using the doubly censored mean function m(x, y) = E(X | x⩽X⩽y). We show that m(x, y) characterizes F(x), obtaining the explicit expression of F(x) from m(x, y). Moreover, we give properties that any function must verifies to be a doubly censored mean function and we obtain stability theorems for these characterizations.     

Nonparametric statistical procedures for the changepoint problem

Let X₁,…,X_r?1,X_r,X_r+1,…,X_n be independent, continuous random variables such that X_i, i = 1,…,r, has distribution function F(x), and X_i, i = r+1,…,n, has distribution function F(x?Δ), with -∞ <Δ< ∞. When the integer r is unknown, this is refered to as a change point problem with at most one change. The unknown parameter Δ represents the magnitude of the change and r is called the changepoint. In this paper we present a general review discussion of several nonparametric approaches for making inferences about r and Δ.     

Limiting distributions of Kolmogorov-Lévy-type statistics under the alternative

Let X_i, 1 ≤ i ≤ n, be independent identically distributed random variables with a common distribution function F, and let G be a smooth distribution function. We derive the limit distribution of □{ρ_α(F_n, G) - _α(F, G)}, where F_n is the empirical distribution function based on X₁,…,X_n and _α is a Kolmogorov-Lévy-type metric between distribution functions. For α ≤ 0 and two distribution functions F and G the metric p_α is given by p_α(F, G) = inf {? ≤ 0: G(x - α?) - ? F(x) ≤ G(x + α?) + ? for all x ?}.     

A characterization of some {3v1 + v3, 3v0 + v2; 3, 3}-minihypers and its applications to error-correcting codes

Recently, Hamada et al. (Des. Codes Cryptogr. 2(1992), 225–229) showed that there exists an [87,5,57;3]-code meeting the Grismer bound. But it is unknown whether or not an [87,5,57;3]-code is unique up to equivalence. In order to characterize all [87,5,57;3]-codes, it is sufficient to characterize all {2v₂ + 2v₃,2v₁ + 2v₂; 4, 3}-minihypers and in order to characterize all {2v₂ + 2v₃,2v₁ + 2v₂;4,3}-minihypers, it is necessary to characterize all {3v₁ + v₃,3v₀ + v₂;3,3}-minihypers, where v₀ = 0, v₁ = 1, v₂ = 4 and v₃ = 13 (cf. Hamada and Helleseth, J. Statist. Plann. Inference, 56 (1996)). The purpose of this paper is to characterize all {3v₁ + v₃,3v₀ + v₂;3,3}-minihypers.     

Lois limites pour les statistiques d'ordre dans le cas non identiquement distribue

Let X₁, X₂,… be a sequence of independent random variables with distribution functions F₁, where 1 ≤ i ≤ n, and for each n ≥ 1 let X_1,n ≤… ≤ X_n,n denote the order statistics of the first n random variables. Under suitable hypotheses about the F₁, we characterize the limit distribution functions H(x) for which P(X_k,n ? a_nx + b_n) → H(x), where a_n > 0 and b_n are real constants. We consider the cases where κ = κ(n) satisfies √n {κ(n)/n — λ} → 0 and √n {κ(n)/n — λ} → ∞ separately.     

Asymptotic Properties of Random Weighted Empirical Distribution Function

This article studies the asymptotic properties of the random weighted empirical distribution function of independent random variables. Suppose X₁, X₂, ???, X_n is a sequence of independent random variables, and this sequence is not required to be identically distributed. Denote the empirical distribution function of the sequence by F_n(x). Based on the random weighting method and F_n(x), the random weighted empirical distribution function H_n(x) is constructed and the asymptotic properties of H_n are discussed. Under weak conditions, the Glivenko–Cantelli theorem and the central limit theorem for the random weighted empirical distribution function are obtained. The obtained results have also been applied to study the distribution functions of random errors of multiple sensors.     

A formula for P(bi≤Ri≤ai, 1≤i≤m) under a general class of alternatives

Let X_i≤?≤X_m and Y_i≤?≤Y_n be two sets of independent order statistics from continous distributions with distribution functions F and G respectively. Let R_i denote the rank of X_i in the combined order sample. Steck (1980) has found an expression for P(b_i≤R_i≤a_i, all i) when F = h(G), h being the incomplete beta function with parameters (α,β?α+1). An alternative expression for the same probability is obtained which is computationally a substantial improvement on Steck's result.     

BAHADUR SLOPES FOR COMPARISONS OF TESTS BASED ON RESAMPLING

Let X₁,…, X_n be random variables symmetric about θ from a common unknown distribution F_θ(x) =F(x–θ). To test the null hypothesis H₀:θ= 0 against the alternative H₁:θ > 0, permutation tests can be used at the cost of computational difficulties. This paper investigates alternative tests that are computationally simpler, notably some bootstrap tests which are compared with permutation tests. Of these the symmetrical bootstrap-f test competes very favourably with the permutation test in terms of Bahadur asymptotic efficiency, so it is a very attractive alternative.     

An optimal search procedure

Consider the following problem. There are exactly two defective (unknown) elements in the set X={x₁, x₂,…,x_n}, all possibilities occuring with equal probabilities. We want to identify the unknown (defective) elements by testing some subsets A of X, and for each such set A determining whether A contains any of them. The test on an individual subset A informs us that either all elements of the tested set A are good, or that at least one of them is defective (but we do not know which ones or how many). A set containing at least one defective element is said to be defective. Our aim is to minimize the maximal number of tests. For the optimal strategy, let the maximal test length be denoted by l₂(n). We obtain the value of this function for an infinite sequence of values of n.     

Moments of Order Statistics from Weibull Distribution in the Presence of Multiple Outliers

In this article, we derive exact expressions for the single and product moments of order statistics from Weibull distribution under the contamination model. We assume that X₁, X₂, …, X_{n ? p} are independent with density function f(x) while the remaining, p observations (outliers) X_{n ? p + 1}, …, X_n are independent with density function arises from some modified version of f(x), which is called g(x), in which the location and/or scale parameters have been shifted in value. Next, we investigate the effect of the outliers on the BLUE of the scale parameter. Finally, we deduce some special cases.     

A characterization of some {3v2 + v3, 3v1 + v2; 3, 3}-minihypers and some [15, 4, 9; 3]-codes with B2 = 0

It is known (cf. Hill and Newton (Ars Combin., 25A (1988), 61–72; Designs Codes Cryptography, 2 (1992), 137–157) and Remark A.2 in the Appendix) that (1) there is no [14, 4, 9; 3]-code meeting the Griesmer bound and (2) if C is a [15, 4, 9; 3]-code then B₂ = 0 or 2 and (3) there is a one-to-one correspondence between the set of all nonequivalent [15, 4, 9; 3]-codes with B₂ = 0 and the set of all {3v₂ + v₃, 3v₁ + v₂: 3, 3}-minihypers, where v₁ = 1, v₂ = 4, v₃ = 13 and B₂ denotes the number of codewords of weight 2 in its dual code. Recently it has been shown by Eupen and Hill (Designs Codes Cryptography, 4 (1994) 271–282) that a [15, 4, 9; 3]-code with B₂ = 2 is unique up to equivalence. The purpose of this paper is to characterize all [15, 4, 9; 3]-codes with B₂ = 0 using the geometrical structure of the corresponding {3v₂ + v₃, 3v₁ + v₂; 3, 3}-minihypers. Those results give a complete characterization of [15, 4, 9; 3]-codes.     

The q-Bernstein basis as a q-binomial distribution

The q-Bernstein basis, used in the definition of the q-Bernstein polynomials, is shown to be the probability mass function of a q-binomial distribution. This distribution is defined on a sequence of zero–one Bernoulli trials with probability of failure at any trial increasing geometrically with the number of previous failures. A modification of this model, with the probability of failure at any trial decreasing geometrically with the number of previous failures, leads to a second q-binomial distribution that is also connected to the q-Bernstein polynomials. The q-factorial moments as well as the usual factorial moments of these distributions are derived. Further, the q-Bernstein polynomial B_n(f(t),q;x) is expressed as the expected value of the function f([X_n]_q/[n]_q) of the random variable X_n obeying the q-binomial distribution. Also, using the expression of the q-moments of X_n, an explicit expression of the q-Bernstein polynomial B_n(f_r(t),q;x), for f_r(t) a polynomial, is obtained.     

Density estimation for samples satisfying a certain absolute regularity condition

Let {ξ_i} be an absolutely regular sequence of identically distributed random variables having common density function f(x). Let H_k(x,y) (k=1, 2,…) be a sequence of Borel-measurable functions and f_n(x)=n^?1(H_n(x,ξ₁)+…+H_n(x,ξ_n)) the empirical density function. In this paper, the asymptotic property of the probability P(sup_x|f_n(x)?f(x)|>ε) (n→∞) is studied.