On the accuracy of empirical likelihood confidence regions for linear regression model

The coverage errors of the empirical likelihood confidence regions for in a linear regression model,Y _i=x _i+ _i , 1in, are of ordern ^–1. Bartlett corrections may be employed to reduce the order of magnitude of the coverage errors ton ^–2. For practical implementation of Bartlett correction, an empirical Bartlett correction is given.     

Smoothing noisy data with spline functions

Summary A procedure for calculating the trace of the influence matrix associated with a polynomial smoothing spline of degree2m–1 fitted ton distinct, not necessarily equally spaced or uniformly weighted, data points is presented. The procedure requires orderm ² n operations and therefore permits efficient orderm ² n calculation of statistics associated with a polynomial smoothing spline, including the generalized cross validation. The method is a significant improvement over an existing method which requires ordern ³ operations.     

Unusual properties of bootstrap confidence intervals in regression problems

Summary We show that the percentile-t method, and one of the two percentile methods, have unusually good performance when employed to construct bootstrap confidence intervals in a regression setting. In the case of slope parameters, percentile-t produces two-sided intervals with coverage errorn ^-2, and one-sided intervals with coverage errorn ^-3/2, wheren is sample size. The errors are onlyn ^-1 in most other problems. One of the percentile methods produces critical points which are third-order correct for Efron's [11] relatively complex accelerated bias-corrected points.     

The asymptotics of integral functionals of a multidimensional empirical process

Estimates of a rate of convergence of orderO(n ^–1) are obtained for integral functionals of an empirical process.Translated fromTeoriya Sluchainykh Protsessov, Vol. 15, pp. 10–16, 1987.     

Reduktionsverfahren für Differenzengleichungen bei Randwertaufgaben II

Summary In a recent paper [11], two of the authors investigated a fast reduction method for solving difference equations which approximate certain boundary value problems for Poisson's equation. In this second paper, we prove the numerical stability of the reduction method, and also report on further developments of the method. For the general case, the provided bounds for the numerical errors behave roughly like the condition numberO(n ²) of the linear system; for more realistic model problems estimates of order less thanO(n) are obtained (n ^–1=h=mesh width). The number of operations required for the reduction method isO(n ² ), for the usual five-point difference formula, as well as for the common nine-point formula with discretization error of orderh ⁴.     

An efficient method for calculating smoothing splines using orthogonal transformations

Summary A procedure for calculating the mean squared residual and the trace of the influence matrix associated with a polynomial smoothing spline of degree 2m–1 using an orthogonal factorization is presented. The procedure substantially overcomes the problem of ill-conditioning encountered by a recently developed method which employs a Cholesky factorization, but still requires only orderm ² n operations and ordermn storage.     

Modeling the coverage probability of a confidence interval on the among-group variance component in the unbalanced random one-way model

The coverage probability of an approximate confidence interval on the among-group variance component, σ _α ² , in a one-way random model is modeled using generalized linear models techniques. The purpose of the proposed modeling is to derive an empirical relationship between the coverage probability on one hand, andk, n., ?, and the model’s variance components on the other hand, wherek is the number of groups, n. is the total number of observations, and ? is a measure of imbalance for the design used. The latter quantities serve as control variables in the derived model, and the coverage probability is treated as the response variable. Contour plots generated from this model can be easily used to depict the effects of the control variables on the coverage probability. In particular, the plots are utilized to compare four methods for constructing approximate confidence intervals on σ _α ² . Additional advantages of the derived model include prediction of the coverage probability for a given method using only specified values of the control variables, and the determination of operating conditions that result in improved coverage probability within the region of interest.     

The product-limit estimator and the bootstrap: Some asymptotic representations

Summary The product-limit estimator and its quantile process are represented as i.i.d. mean processes, with a remainder of ordern ^–3/4(logn)^3/4 a.s. Corresponding bootstrap versions of these representations are given, which can help one visualize how the bootstrap procedure operates in this set up.Research supported by NSF grants MCS-81-02341 and MCS 83-01082     

A smoothed bootstrap estimator for a studentized sample quantile

If the underlying distribution functionF is smooth it is known that the convergence rate of the standard bootstrap quantile estimator can be improved fromn ^–1/4 ton ^–1/2+, for arbitrary >0, by using a smoothed bootstrap. We show that a further significant improvement of this rate is achieved by studentizing by means of a kernel density estimate. As a consequence, it turns out that the smoothed bootstrap percentile-t method produces confidence intervals with critical points being second-order correct and having smaller length than competitors based on hybrid or on backwards critical points. Moreover, the percentile-t method for constructing one-sided or two-sided confidence intervals leads to coverage accuracies of ordern ^–1+, for arbitrary >0, in the case of analytic distribution functions.     

Towards a characterization of generalized twisted field planes

In this work we study semifield planes of orderq ⁿ(q=p ^r ,p prime) with a collineation whose order is ap-primitive divisor ofq ⁿ–1.Research supported in part by NSF Grant No. DMS-9107372Research supported in part by NSF grants RII-9014056, component IV of the EPSCoR of Puerto Rico grant and ARO grant for Cornell MSI.     

Order Determination for Multivariate Autoregressive Processes Using Resampling Methods

LetX₁, …, X_nbe observations from a multivariate AR(p) model with unknown orderp. A resampling procedure is proposed for estimating the orderp. The classical criteria, such as AIC and BIC, estimate the orderpas the minimizer of the function[formula]wherenis the sample size,kis the order of the fitted model, Σ²_kis an estimate of the white noise covariance matrix, andC_nis a sequence of specified constants (for AIC,C_n=2m²/n, for Hannan and Quinn's modification of BIC,C_n=2m²(ln ln n)/n, wheremis the dimension of the data vector). A resampling scheme is proposed to estimate an improved penalty factorC_n. Conditional on the data, this procedure produces a consistent estimate ofp. Simulation results support the effectiveness of this procedure when compared with some of the traditional order selection criteria. Comments are also made on the use of Yule–Walker as opposed to conditional least squares estimations for order selection.     

Asymptotic expansion formula for Bernstein polynomials defined on a simplex

In this paper we give a complete expansion formula for Bernstein polynomials defined on ans-dimensional simplex. This expansion for a smooth functionf represents the Bernstein polynomialB _n (f) as a combination of derivatives off plus an error term of orderO(n^–s ).Communicated by Wolfgang Dahmen.     

Backward differentiation formulas with extended regions of absolute stability

This paper presents a class of (p + 2)-step backward differentiation formulas of orderp. The two extra degrees of freedom obtained by limiting the order of a (p + 2)-step formula top are used to extend the region of absolute stability. A new formula of orderp has a region of absolute stability very similar to that of a classical backward differentiation formula of orderp - 1 forp being in the range 4–6. The backward differentiation formulas with extended regions of absolute stability are constructed by appending two exponential-trigonometric terms to the polynomial basis of the classical formulas. Besides the absolute stability, the paper discusses relative stability and contractivity. The principles of an experimental implementation of the new formulas are outlined, and a linear problem integrated with this computer program indicates that the extended regions of absolute stability can actually be exploited in practice.     

Gaussian integration of Chebyshev polynomials and analytic functions

Explicit bounds for the quadrature error of thenth Gauss-Legendre quadrature rule applied to themth Chebyshev polynomial are derived. They are precise up to the orderO(m ⁴ n ^–6). As an application, error constants for classes of functions which are analytic in the interior of an ellipse are estimated. The location of the maxima of the corresponding kernel function is investigated.Dedicated to Luigi Gatteschi on the occasion of his 70th birthday     

The structure of weighing matrices having large weights

We examine the structure of weighing matricesW(n, w), wherew=n–2,n–3,n–4, obtaining analogues of some useful results known for the casen–1. In this setting we find some natural applications for the theory ofsigned groups and orthogonal matrices with entries from signed groups, as developed in [3]. We construct some new series of Hadamard matrices from weighing matrices, including the following:W(n, n–2) implies an Hadamard matrix of order2n ifn0 mod 4 and order 4n otherwise;W(n, n–3) implies an Hadamard matrix of order 8n; in certain cases,W(n, n–4) implies an Hadamard matrix of order 16n. We explicitly derive 117 new Hadamard matrices of order 2 ^t p, p<4000, the smallest of which is of order 2³·419.Supported by an NSERC grant     

Constructing cospectral graphs

Some new constructions for families of cospectral graphs are derived, and some old ones are considerably generalized. One of our new constructions is sufficiently powerful to produce an estimated 72% of the 51039 graphs on 9 vertices which do not have unique spectrum. In fact, the number of graphs of ordern without unique spectrum is believed to be at leastn ³ g _–1 for some>0, whereg _n is the number of graphs of ordern andn 7.     

Escape probabilities for slowly recurrent sets

Summary A setAZ ^d (d>-3) is defined to be slowly recurrent for simple random walk if it is recurrent but the probability of enteringA{z:n<|z|<-2n} tends to zero asn. A method is given to estimate escape probabilities for such sets, i.e., the probability of leaving the ball of radiusn without entering the set. The methods are applied to two examples. First, half-lines and finite unions of half-lines inZ ³ are considered. The second example is a random walk path in four dimensions. In the latter case it is proved that the probability that two random walk paths reach the ball of radiusn without intersecting is asymptotic toc(lnn)^–1/2, improving a result of the author.Research partially supported by the National Science Foundation     

Lower bounds for bandwidth selection in density estimation

Summary This paper establishes asymptotic lower bounds which specify, in a variety of contexts, how well (in terms of relative rate of convergence) one may select the bandwidth of a kernel density estimator. These results provide important new insights concerning how the bandwidth selection problem should be considered. In particular it is shown that if the error criterion is Integrated Squared Error (ISE) then, even under very strong assumptions on the underlying density, relative error of the selected bandwidth cannot be reduced below ordern ^–1/10 (as the sample size grows). This very large error indicates that any technique which aims specifically to minimize ISE will be subject to serious practical difficulties arising from sampling fluctuations. Cross-validation exhibits this very slow convergence rate, and does suffer from unacceptably large sampling variation. On the other hand, if the error criterion is Mean Integrated Squared Error (MISE) then relative error of bandwidth selection can be reduced to ordern ^–1/2, when enough smoothness is assumed. Therefore bandwidth selection techniques which aim to minimize MISE can be much more stable, and less sensitive to small sampling fluctuations, than those which try to minimize ISE. We feel this indicates that performance in minimizing MISE, rather than ISE, should become the benchmark for measuring performance of bandwidth selection methods.Research partially supported by National Science Foundation Grants DMS-8701201 and DMS-8902973Research of the first author was done while on leave from the Australian National University     

Base tensorielle des matrices de Hankel (ou de Toeplitz) Applications

Summary The set of all Hankel (or Toeplitz) matrices of dimensionn, is shown to possess tensorial bases: bases made ofn rank one matrices. Four families of such tensorial bases are possible. From this result, we deduce that the following computations can be performed with a number of multiplications of ordern instead of ordern ²: evaluation of the 2n+1 coefficients of the polynomial product of two polynomials of degreen, evaluation of the inverse of a lower triangular toeplitz matrix, evaluation of the quotient and of the remainder in the division of a polynomial of degree 2n by a polynomial of degreen.     

Deriving the beta-constraints for geometric continuity of parametric curves

Parametric splines curves are typically constructed so that the firstn parametric derivatives agree where the curve segments abut. This type of continuity condition has become known asC ⁿ orn ^th orderparametric continuity. It has previously been shown that the use of parametric continuity disallows many parametrizations which generate geometrically smooth curves. We definen ^th ordergeometric continuity (Gⁿ), develop constraint equations that are necessary and sufficient for geometric continuity of curves, and show that geometric continuity is a relaxed form of parametric continuity.G ⁿ continuity provides for the introduction ofn quantities known asshape parameters which can be made available to a designer in a computer aided design environment to modify the shape of curves without moving control vertices. Several applications of the theory are discussed, along with topics of future research.