Automatic Smoothing With Wavelets for a Wide Class of Distributions

Wavelet-based denoising techniques are well suited to estimate spatially inhomogeneous signals. Waveshrink (Donoho and Johnstone) assumes independent Gaussian errors and equispaced sampling of the signal. Various articles have relaxed some of these assumptions, but a systematic generalization to distributions such as Poisson, binomial, or Bernoulli is missing. We consider a unifying l₁-penalized likelihood approach to regularize the maximum likelihood estimation by adding an l₁ penalty of the wavelet coefficients. Our approach works for all types of wavelets and for a range of noise distributions. We develop both an algorithm to solve the estimation problem and rules to select the smoothing parameter automatically. In particular, using results from Poisson processes, we give an explicit formula for the universal smoothing parameter to denoise Poisson measurements. Simulations show that the procedure is an improvement over other methods. An astronomy example is given.     

Boosting for Correlated Binary Classification

Boosting is a successful method for dealing with problems of high-dimensional classification of independent data. However, existing variants do not address the correlations in the context of longitudinal or cluster study-designs with measurements collected across two or more time points or in clusters. This article presents two new variants of boosting with a focus on high-dimensional classification problems with matched-pair binary responses or, more generally, any correlated binary responses. The first method is based on the generic functional gradient descent algorithm and the second method is based on a direct likelihood optimization approach. The performance and the computational requirements of the algorithms were evaluated using simulations. Whereas the performance of the two methods is similar, the computational efficiency of the generic-functional-gradient-descent-based algorithm far exceeds that of the direct-likelihood-optimization-based algorithm. The former method is illustrated using data on gene expression changes in de novo and relapsed childhood acute lymphoblastic leukemia. Computer code implementing the algorithms and the relevant dataset are available online as supplemental materials.     

Weighted Generalized Inverses,Oblique Projections,and Least-Squares Problems

A generalization with singular weights of Moore–Penrose generalized inverses of closed range operators in Hilbert spaces is studied using the notion of compatibility of subspaces and positive operators.     

Nikolskii type inequality for cardinal splines

The Nikolskii type inequality for cardinal splines

Collocation by singular splines

Splines determined by the kernel of the differential operator are known to be useful to solve the singular boundary value problems of the form . One of the most successful methods is the collocation method based on special Chebyshev splines. We investigate the construction of the associated B-splines based on knot-insertion algorithms for their evaluation, and their application in collocation at generalized Gaussian points. Specially, we show how to obtain these points as eigenvalues of a symmetric tridiagonal matrix of order k. This research was supported by Grant 037-1193086-2771, by the Ministry of science, education and sports of the Republic of Croatia.     

Efficient Spatial Modeling Using the SPDE Approach With Bivariate Splines

Gaussian fields (GFs) are frequently used in spatial statistics for their versatility. The associated computational cost can be a bottleneck, especially in realistic applications. It has been shown that computational efficiency can be gained by doing the computations using Gaussian Markov random fields (GMRFs) as the GFs can be seen as weak solutions to corresponding stochastic partial differential equations (SPDEs) using piecewise linear finite elements. We introduce a new class of representations of GFs with bivariate splines instead of finite elements. This allows an easier implementation of piecewise polynomial representations of various degrees. It leads to GMRFs that can be inferred efficiently and can be easily extended to nonstationary fields. The solutions approximated with higher order bivariate splines converge faster, hence the computational cost can be alleviated. Numerical simulations using both real and simulated data also demonstrate that our framework increases the flexibility and efficiency. Supplementary materials are available online.     

动态变系数回归模型的识别与非参数GMM估计

主要研究关于面板数据的有限阶固定效应的动态变系数回归模型(简称FDVCM)的统计推断问题.基于B-样条函数和广义矩估计(简称GMM)方法,首先建立了未知系数函数的非参数GMM估计,并证明大样本情形下该估计达到最优非参数收敛速度且具有渐近正态性质.然而实际问题中模型的动态阶数完全未知,也可能存在其它冗余的回归变量,文中借助文[Fan J,Li R.Variable selection via penalized likelihood and its oracle properties.Journal of the American Statistical Association,2001,96(456):1348-1360]中的smoothly clipped absolute deviation(简称SCAD)惩罚函数同时识别真实的动态阶数和显著的外生回归变量.同时建立了压缩估计的Oracle性质,即所识别的模型与真实模型中的参数估计具有相同的渐近分布.最后,无论是数值试验还是实例数据分析都验证了本文方法的合理性和可行性.     

The convergence of three spline methods for scattered data interpolation and fitting

The convergences of three L₁ spline methods for scattered data interpolation and fitting using bivariate spline spaces are studied in this paper. That is, L₁ interpolatory splines, splines of least absolute deviation, and L₁ smoothing splines are shown to converge to the given data function under some conditions and hence, the surfaces from these three methods will resemble the given data values.     

Fast evaluation of polyharmonic splines in three dimensions

M. J. D. Powell ^{This paper concerns the fast evaluation of radial basis functions.It describes the mathematics of hierarchical and fast multipolemethods for fast evaluation of splines of the form where is a positive integer andp is a low-degree polynomial. Splines s of this form are polyharmonicsplines in 3 and have been found to be very useful for providingsolutions to scattered data interpolation problems in 3. Asit is now well known, hierarchical methods reduce the incrementalcost of a single extra evaluation from O(N) to O(log N) operationsand reduce the cost of a matrix–vector product (evaluationof s at all the centres) from O(N2) to O(N log N) operations.We give appropriate far- and near-field expansions, togetherwith error estimates, uniqueness theorems and translation formulae.A hierarchical code based on these formulae is detailed andsome numerical results are given.}