Searcharxiv⌕ Search

arXiv subjects

Larry Goldstein

Publications and source records attributed to Larry Goldstein.

51 records · Page 3Linked to original sources
arXiv2010

The Spend-It-All Region and Small Time Results for the Continuous Bomber Problem

A problem of optimally allocating partially effective ammunition $x$ to be used on randomly arriving enemies in order to maximize an aircraft's probability of surviving for time~$t$, known as the Bomber Problem, was first posed by \citet{Klinger68}. They conjectured a set of apparently obvious monotonicity properties of the optimal allocation function $K(x,t)$. Although some of these conjectures, and versions thereof, have been proved or disproved by other authors since then, the remaining central question, that $K(x,t)$ is nondecreasing in~$x$, remains unsettled. After reviewing the problem and summarizing the state of these conjectures, in the setting where $x$ is continuous we prove the existence of a ``spend-it-all'' region in which $K(x,t)=x$ and find its boundary, inside of which the long-standing, unproven conjecture of monotonicity of~$K(\cdot,t)$ holds. A new approach is then taken of directly estimating~$K(x,t)$ for small~$t$, providing a complete small-$t$ asymptotic description of~$K(x,t)$ and the optimal probability of survival.

math.PR↗
arXiv2009

Efficiency of the maximum partial likelihood estimator for nested case control sampling

In making inference on the relation between failure and exposure histories in the Cox semiparametric model, the maximum partial likelihood estimator (MPLE) of the finite dimensional odds parameter, and the Breslow estimator of the baseline survival function, are known to achieve full efficiency when data is available for all time on all cohort members, even when the covariates are time dependent. When cohort sizes become too large for the collection of complete data, sampling schemes such as nested case control sampling must be used and, under various models, there exist estimators based on the same information as the MPLE having smaller asymptotic variance. Though the MPLE is therefore not efficient under sampling in general, it approaches efficiency in highly stratified situations, or instances where the covariate values are increasingly less dependent upon the past, when the covariate distribution, not depending on the real parameter of interest, is unknown and there is no censoring. In particular, in such situations, when using the nested case control sampling design, both the MPLE and the Breslow estimator of the baseline survival function achieve the information lower bound both in the distributional and the minimax senses in the limit as the number of cohort members tends to infinity.

math.ST↗
arXiv2007

$L^1$ bounds in normal approximation

The zero bias distribution $W^*$ of $W$, defined though the characterizing equation $\mathit{EW}f(W)=σ^2Ef'(W^*)$ for all smooth functions $f$, exists for all $W$ with mean zero and finite variance $σ^2$. For $W$ and $W^*$ defined on the same probability space, the $L^1$ distance between $F$, the distribution function of $W$ with $\mathit{EW}=0$ and $Var(W)=1$, and the cumulative standard normal $Φ$ has the simple upper bound \[\Vert F-Φ\Vert_1\le2E|W^*-W|.\] This inequality is used to provide explicit $L^1$ bounds with moderate-sized constants for independent sums, projections of cone measure on the sphere $S(\ell_n^p)$, simple random sampling and combinatorial central limit theorems.

math.PR↗
arXiv2006

Zero biasing and a discrete central limit theorem

We introduce a new family of distributions to approximate $\mathbb {P}(W\in A)$ for $A\subset\{...,-2,-1,0,1,2,...\}$ and $W$ a sum of independent integer-valued random variables $ξ_1$, $ξ_2$, $...,$ $ξ_n$ with finite second moments, where, with large probability, $W$ is not concentrated on a lattice of span greater than 1. The well-known Berry--Esseen theorem states that, for $Z$ a normal random variable with mean $\mathbb {E}(W)$ and variance $\operatorname {Var}(W)$, $\mathbb {P}(Z\in A)$ provides a good approximation to $\mathbb {P}(W\in A)$ for $A$ of the form $(-\infty,x]$. However, for more general $A$, such as the set of all even numbers, the normal approximation becomes unsatisfactory and it is desirable to have an appropriate discrete, nonnormal distribution which approximates $W$ in total variation, and a discrete version of the Berry--Esseen theorem to bound the error. In this paper, using the concept of zero biasing for discrete random variables (cf. Goldstein and Reinert [J. Theoret. Probab. 18 (2005) 237--260]), we introduce a new family of discrete distributions and provide a discrete version of the Berry--Esseen theorem showing how members of the family approximate the distribution of a sum $W$ of integer-valued variables in total variation.

math.PR↗
arXiv2006

Cohort sampling schemes for the Mantel-Haenszel estimator: Extensions to multilevel covariates, stratified models, and robust variance estimators

In many epidemiological contexts, disease occurrences and their rates are naturally modelled by counting processes and their intensities, allowing an analysis based on martingale methods. These methods lend themselves to extensions of nested case-control sampling designs where general methods of control selection can be easily incorporated. This same methodology allows for extensions of the Mantel-Haenszel estimator in two main directions. First, a variety of new sampling designs can be incorporated which can yield substantial efficiency gains over simple random sampling. Second, the extension allows for the treatment of multiple level time dependent exposures.

math.ST↗
arXiv2005

Berry Esseen bounds for combinatorial central limit theorems and pattern occurrences, using zero and size biasing

Berry Esseen type bounds to the normal, based on zero- and size-bias couplings, are derived using Stein's method. The zero biasing bounds are illustrated with an application to combinatorial central limit theorems where the random permutation has either the uniform distribution or one which is constant over permutations with the same cycle type and having no fixed points. The size biasing bounds are applied to the occurrences of fixed relatively ordered sub-sequences (such as rising sequences) in a random permutation, and to the occurrences of patterns, extreme values, and subgraphs on finite graphs.

math.PR↗
arXiv2005

A permutation test for matching and its asymptotic distribution

We consider a permutation method for testing whether observations given in their natural pairing exhibit an unusual level of similarity in situations where any two observations may be similar at some unknown baseline level. Under a null hypotheses where there is no distinguished pairing of the observations, a normal approximation with explicit bounds and rates is presented for determining approximate critical test levels.

math.ST↗
arXiv2005

Stein's method and the zero bias transformation with application to simple random sampling

Let $W$ be a random variable with mean zero and variance $σ^2$. The distribution of a variate $W^*$, satisfying $EWf(W)=σ^2 Ef'(W^*)$ for smooth functions $f$, exists uniquely and defines the zero bias transformation on the distribution of $W$. The zero bias transformation shares many interesting properties with the well known size bias transformation for non-negative variables, but is applied to variables taking on both positive and negative values. The transformation can also be defined on more general random objects. The relation between the transformation and the expression $wf'(w)-σ^2 f''(w)$ which appears in the Stein equation characterizing the mean zero, variance $σ^2$ normal $σZ$ can be used to obtain bounds on the difference $E\{h(W/σ)-h(Z)\}$ for smooth functions $h$ by constructing the pair $(W,W^*)$ jointly on the same space. When $W$ is a sum of $n$ not necessarily independent variates, under certain conditions which include a vanishing third moment, bounds on this difference of the order $1/n$ for classes of smooth functions $h$ may be obtained. The technique is illustrated by an application to simple random sampling.

math.PR↗
arXiv2005

Multivariate normal approximations by Stein's method and size bias couplings

Stein's method is used to obtain two theorems on multivariate normal approximation. Our main theorem, Theorem 1.2, provides a bound on the distance to normality for any nonnegative random vector. Theorem 1.2 requires multivariate size bias coupling, which we discuss in studying the approximation of distributions of sums of dependent random vectors. In the univariate case, we briefly illustrate this approach for certain sums of nonlinear functions of multivariate normal variables. As a second illustration, we show that the multivariate distribution counting the number of vertices with given degrees in certain random graphs is asymptotically multivariate normal and obtain a bound on the rate of convergence. Both examples demonstrate that this approach may be suitable for situations involving non-local dependence. We also present Theorem 1.4 for sums of vectors having a local type of dependence. We apply this theorem to obtain a multivariate normal approximation for the distribution of the random $p$-vector which counts the number of edges in a fixed graph both of whose vertices have the same given color when each vertex is colored by one of $p$ colors independently. All normal approximation results presented here do not require an ordering of the summands related to the dependence structure. This is in contrast to hypotheses of classical central limit theorems and examples, which involve e.g., martingale, Markov chain, or various mixing assumptions.

math.PR↗
arXiv2005

An unexpected connection between branching processes and optimal stopping

A curious connection exists between the theory of optimal stopping for independent random variables, and branching processes. In particular, for the branching process $Z_n$ with offspring distribution $Y$, there exists a random variable $X$ such that the probability $P(Z_n=0)$ of extinction of the $n$th generation in the branching process equals the value obtained by optimally stopping the sequence $X_1,...,X_n$, where these variables are i.i.d distributed as $X$. Generalizations to the inhomogeneous and infinite horizon cases are also considered. This correspondence furnishes a simple `stopping rule' method for computing various characteristics of branching processes, including rates of convergence of the $n^{th}$ generation's extinction probability to the eventual extinction probability, for the supercritical, critical and subcritical Galton-Watson process. Examples, bounds, further generalizations and a connection to classical prophet inequalities are presented. Throughout, the aim is to show how this unexpected connection can be used to translate methods from one area of applied probability to another, rather than to provide the most general results.

math.PR↗
arXiv2005

Distributional transformations, orthogonal polynomials, and Stein characterizations

A new class of distributional transformations is introduced, characterized by equations relating function weighted expectations of test functions on a given distribution to expectations of the transformed distribution on the test function's higher order derivatives. The class includes the size and zero bias transformations, and when specializing to weighting by polynomial functions, relates distributional families closed under independent addition, and in particular the infinitely divisible distributions, to the family of transformations induced by their associated orthogonal polynomial systems. For these families, generalizing a well known property of size biasing, sums of independent variables are transformed by replacing summands chosen according to a multivariate distribution on its index set by independent variables whose distributions are transformed by members of that same family. A variety of the transformations associated with the classical orthogonal polynomial systems have as fixed points the original distribution, or a member of the same family with different parameter.

math.PR↗
arXiv2005

Two choice optimal stopping

Let $X_n,...,X_1$ be i.i.d. random variables with distribution function $F$. A statistician, knowing $F$, observes the $X$ values sequentially and is given two chances to choose $X$'s using stopping rules. The statistician's goal is to stop at a value of $X$ as small as possible. Let $V_n^2$ equal the expectation of the smaller of the two values chosen by the statistician when proceeding optimally. We obtain the asymptotic behavior of the sequence $V_n^2$ for a large class of $F$'s belonging to the domain of attraction (for the minimum) ${\cal D}(G^α)$, where $G^α(x)=[1-\exp(-x^α)]{\bf I}(x \ge 0)$. The results are compared with those for the asymptotic behavior of the classical one choice value sequence $V_n^1$, as well as with the ``prophet value" sequence $V_n^p=E(\min\{X_n,...,X_1\})$.

math.PR↗
arXiv2005

Local central limit theorems, the high-order correlations of rejective sampling and logistic likelihood asymptotics

Let I_1,...,I_n be independent but not necessarily identically distributed Bernoulli random variables, and let X_n=\sum_{j=1}^nI_j. For νin a bounded region, a local central limit theorem expansion of P(X_n=EX_n+ν) is developed to any given degree. By conditioning, this expansion provides information on the high-order correlation structure of dependent, weighted sampling schemes of a population E (a special case of which is simple random sampling), where a set d\subset E is sampled with probability proportional to \prod_{A\in d}x_A, where x_A are positive weights associated with individuals A\in E. These results are used to determine the asymptotic information, and demonstrate the consistency and asymptotic normality of the conditional and unconditional logistic likelihood estimator for unmatched case-control study designs in which sets of controls of the same size are sampled with equal probability.

math.ST↗
arXiv2005

Normal approximation for hierarchical structures

Given F:[a,b]^k\to [a,b] and a nonconstant X_0 with P(X_0\in [a,b])=1, define the hierarchical sequence of random variables {X_n}_{n\ge 0} by X_{n+1}=F(X_{n,1},...,X_{n,k}), where X_{n,i} are i.i.d. as X_n. Such sequences arise from hierarchical structures which have been extensively studied in the physics literature to model, for example, the conductivity of a random medium. Under an averaging and smoothness condition on nontrivial F, an upper bound of the form Cγ^n for 0<γ<1 is obtained on the Wasserstein distance between the standardized distribution of X_n and the normal. The results apply, for instance, to random resistor networks and, introducing the notion of strict averaging, to hierarchical sequences generated by certain compositions. As an illustration, upper bounds on the rate of convergence to the normal are derived for the hierarchical sequence generated by the weighted diamond lattice which is shown to exhibit a full range of convergence rate behavior.

math.PR↗