Lower Rank Approximation of Matrices Based on Fast and Robust Alternating Regression

We introduce fast and robust algorithms for lower rank approximation to given matrices based on robust alternating regression. The alternating least squares regression, also called criss-cross regression, was used for lower rank approximation of matrices, but it lacks robustness against outliers in these matrices. We use robust regression estimators and address some of the complications arising from this approach. We find it helpful to use high breakdown estimators in the initial iterations, followed by M estimators with monotone score functions in later iterations towards convergence. In addition to robustness, the computational speed is another important consideration in the development of our proposed algorithm, because alternating robust regression can be computationally intensive for large matrices. Based on a mix of the least trimmed squares (LTS) and Huber's M estimators, we demonstrate that fast and robust lower rank approximations are possible for modestly large matrices.     

利用加权对称估计量对季节性时间序列的单位根检验

本文提出对季节性时间序列利用加权对称估计量的单位根检验,导出相应统计量的极限分布。用MonteCarlo方法计算经验百分位数及检验势,并对最小平方估计量,简单对称估计量和加权对称估计量的经验检验势作了比较。     

Estimating a mean matrix: boosting efficiency by multiple affine shrinkage

The unknown matrix M is the mean of the observed response matrix in a multivariate linear model with independent random errors. This paper constructs regularized estimators of M that dominate, in asymptotic risk, least squares fits to the model and to specified nested submodels. In the first construction, the response matrix is expressed as the sum of orthogonal components determined by the submodels; each component is replaced by an adaptive total least squares fit of possibly lower rank; and these fits are then summed. The second, lower risk, construction differs only in the second step: each orthogonal component is replaced by a modified Efron-Morris fit before summation. Singular value decompositions yield computable formulae for the estimators and their asymptotic and estimated risks. In the asymptotics, the row dimension of M tends to infinity while the column dimension remains fixed. Convergences are uniform when signal-to-noise ratio is bounded. This research was supported in part by National Science Foundation Grant DMS 0404547.     

A robust test of specification based on order statistics

The test of misspecification presented compares parametric and nonparametric regressions. The latter is estimated using order statistics, which provide robust and distribution free estimators. The former is estimated using, in turn, least squares and least absolute deviation (LAD). When implementing LAD, robust and distribution free estimators are considered in both parametric and nonparametric regressions. This defines a very homogeneous test which can discriminate misspecification from the impact of outliers and/or skewness. These two effects are instead mixed together in the tests comparing OLS with nonparametric estimators, potentially driving to erroneous conclusions. An example and a Monte Carlo experiment analyze the behavior of the proposed test.     

Robust Frequency Estimation Using Elemental Sets

Abstract

The extraction of sinusoidal signals from time-series data is a classic problem of ongoing interest in the statistics and signal processing literatures. Obtaining least squares estimates is difficult because the sum of squares has local minima O(1/n) apart in the frequencies. In practice the frequencies are often estimated using ad hoc and inefficient methods. Problems of data quality have received little attention. An elemental set is a subset of the data containing the minimum number of points such that the unknown parameters in the model can be identified. This article shows that, using a variant of the classical method of Prony, parameter estimates for a sum of sinusoids can be obtained algebraically from an elemental set. Elemental set methods are used to construct finite algorithm estimators that approximately minimize the least squares, least trimmed sum of squares, or least median of squares criteria. The elemental set estimators prove able in simulations to resolve the frequencies to the correct local minima of the objective functions. When used as the first stage of an MM estimator, the constructed estimators based on the trimmed sum of squares and least median of squares criteria produce final estimators which have high breakdown properties and which are simultaneously efficient when no outliers are present. The approach can also be applied to sums of exponentials, and sums of damped sinusoids. The article includes simulations with one and two sinusoids and two data examples.     

On nonparametricT-method of multiple,comparisons for randomized blocks

Summary Some nonparametric generalizations of Tukey’s [9]T-method of multiple comparisons are considered for randomized blocks and the allied efficiency results are studied. For this, the distribution theory of aligned rank order statistics developed in [6], [7] is extended for multiple comparisons along the lines of [5] which deals with one-way layouts. Work supported by the National Institute of Health, Public Health Service, Grant GM-12868.     

ASP fits to multi-way layouts

The balanced complete multi-way layout with ordinal or nominal factors is a fundamental data-type that arises in medical imaging, agricultural field trials, DNA microassays, and other settings where analysis of variance (ANOVA) is an established tool. ASP algorithms weigh competing biased fits in order to reduce risk through variance-bias tradeoff. The acronym ASP stands for Adaptive Shrinkage of Penalty bases. Motivating ASP is a penalized least squares criterion that associates a separate quadratic penalty term with each main effect and each interaction in the general ANOVA decomposition of means. The penalty terms express plausible conjecture about the mean function, respecting the difference between ordinal and nominal factors. Multiparametric asymptotics under a probability model and experiments on data elucidate how ASP dominates least squares, sometimes very substantially. ASP estimators for nominal factors recover Stein's superior shrinkage estimators for one- and two-way layouts. ASP estimators for ordinal factors bring out the merits of smoothed fits to multi-way layouts, a topic broached algorithmically in work by Tukey. This research was supported in part by National Science Foundation Grants DMS 0300806 and 0404547.     

广义偏最小二乘(PLS)估计类

本文研究了一类含有偏最小二乘（partialleastsquaresPLS）估计的估计类．给出了PLS估计的一般代数形式；讨论了含有PLS估计的广义PPLS估计类的统计性质；给出了该估计类优于最小二乘估计的条件．     

Influence curves of general statistics

The influence curve (JC) of a Fisher-consistent functional was introduced by F. Hampel and plays a central role in the search for robust estimators. An extension of this notion to non-Fisher-consistent functionals is proposed in order to investigate the infinitesimal robustness of more general statistics, e.g. those used in hypothesis testing. This new definition inherits many useful properties, including some on asymptotic efficiency. Functionals in two variables, arising from two-sample statistics, are treated too. Connections with Hodges-Lehmann shift estimators are discovered. One- and two-sample rank statistics illustrate the theory.     

Bridge estimators and the adaptive lasso under heteroscedasticity

In this paper we investigate penalized least squares methods in linear regression models with heteroscedastic error structure. It is demonstrated that the basic properties with respect to model selection and parameter estimation of bridge estimators, Lasso and adaptive Lasso do not change if the assumption of homoscedasticity is violated. However, these estimators do not have oracle properties in the sense of Fan and Li (2001) if the oracle is based on weighted least squares. In order to address this problem we introduce weighted penalized least squares methods and demonstrate their advantages by asymptotic theory and by means of a simulation study.     

Asymptotics of R-, MD- and LAD-estimators in linear regression models with long range dependent errors

Summary This paper establishes the uniform closeness of a weighted residual empirical process to its natural estimate in the linear regression setting when the errors are Gaussian, or a function of Gaussian random variables, that are strictly stationary and long range dependent. This result is used to yield the asymptotic uniform linearity of a class of rank statistics in linear regression models with long range dependent errors. The latter result, in turn, yields the asymptotic distribution of the Jaeckel (1972) rank estimators. The paper also studies the least absolute deviation and a class of certain minimum distance estimators of regression parameters and the kernel type density estimators of the marginal error density when the errors are long range dependent.Research of this author was partly supported by the NSF grant: DMS-9102041     

Estimation of the variances in the branching process with immigration

Summary Estimation theory for the variances of the offspring and immigration distributions in a simple branching process with immigration is developed, analogous to the estimation theory for the means given by Wei and Winnicki (1990). Conditional and weighted conditional least squares estimators are considered and their asymptotic properties for the full range of parameters are studied. Nonexistence of consistent estimators in the critical case is established, which complements analogous result of Wei and Winnicki for the supercritical case.Research supported by the National Science Foundation under Grant NSF-DMS-8801496     

Asymptotic expansions for the distribution functions of Pickands-type estimators

The asymptotic expansions for the distribution functions of Pickands-type estimators in extreme statistics are obtained. In addition, several useful results on regular variation and intermediate order statistics are presented. Project supported by the National Natural Science Foundation of China (Grant No. 19601007) and Doctoral Program Foundation of Higher Education of China.     

Least median of squares estimation by optimization heuristics with an application to the CAPM and a multi-factor model

For estimating the parameters of models for financial market data, the use of robust techniques is of particular interest. Conditional forecasts, based on the capital asset pricing model, and a factor model are considered. It is proposed to consider least median of squares estimators as one possible alternative to ordinary least squares. Given the complexity of the objective function for the least median of squares estimator, the estimates are obtained by means of optimization heuristics. The performance of two heuristics is compared, namely differential evolution and threshold accepting. It is shown that these methods are well suited to obtain least median of squares estimators for real world problems. Furthermore, it is analyzed to what extent parameter estimates and conditional forecasts differ between the two estimators. The empirical analysis considers daily and monthly data on some stocks from the Dow Jones Industrial Average Index.     

Parameter estimation in two-type continuous-state branching processes with immigration

We study the parameter estimation of two-type continuous-state branching processes with immigration based on low frequency observations at equidistant time points. The ergodicity of the processes is proved. The estimators are based on the minimization of a sum of squared deviation about conditional expectations. We also establish the strong consistency and central limit theorems of the conditional least squares estimators and the weighted conditional least squares estimators of the drift and diffusion coefficients based on low frequency observations.     

Asymptotic properties of the least squares estimators of the parameters of the chirp signals

Chirp signals are quite common in different areas of science and engineering. In this paper we consider the asymptotic properties of the least squares estimators of the parameters of the chirp signals. We obtain the consistency property of the least squares estimators and also obtain the asymptotic distribution under the assumptions that the errors are independent and identically distributed. We also consider the generalized chirp signals and obtain the asymptotic properties of the least squares estimators of the unknown parameters. Finally we perform some simulations experiments to see how the asymptotic results behave for small sample and the performances are quite satisfactory.     

The exponentiated generalized power Lindley distribution: Properties and applications

In this paper, we introduce a new extension of the power Lindley distribution, called the exponentiated generalized power Lindley distribution. Several mathematical properties of the new model such as the shapes of the density and hazard rate functions, the quantile function, moments, mean deviations, Bonferroni and Lorenz curves and order statistics are derived.Moreover, we discuss the parameter estimation of the new distribution using the maximum likelihood and diagonally weighted least squares methods. A simulation study is performed to evaluate the estimators. We use two real data sets to illustrate the applicability of the new model. Empirical findings show that the proposed model provides better fits than some other well-known extensions of Lindley distributions.     

Time-Varying Parameters Prediction

This paper develops a method of adaptive modeling that may be applied to forecast non-stationary time series. The starting point are time-varying coefficients models introduced in statistics, econometrics and engineering. The basic step of modeling is represented by the implementation of adaptive recursive estimators for tracking parameters. This is achieved by unifying basic algorithms—such as recursive least squares (RLS) and extended Kalman filter (EKF)—into a general scheme and next by selecting its coefficients with the minimization of the sum of squared prediction errors. This defines a non-linear estimation problem that may be analyzed in the context of the conditional least squares (CLS) theory. A numerical application on the IBM stock price series of Box-Jenkins illustrates the method and shows its good forecasting ability.     

A unified method to calculate the moments of the least squares estimators in nonlinear regression

A unified method and a set of regularity conditions are presented in this paper to calculate the moments of the least squares estimators, residuals, and fitting errors in nonlinear regression. The results given by Box, Clarke and other authors have been greatly improved and developed. The key point of the method is that we find a series of stochastic expansions related to the estimators. These expansions consist of curvature measures and independent standard normal random variables; therefore they are easy to understand and to deal with from statistical and geometrical point of view.The project is supported by The National Natural Science Foundation of China.     

Transmuted lower record type inverse rayleigh distribution: estimation,characterizations and applications

This study introduces a new lifetime distribution called the transmuted lower record type inverse Rayleigh which extends the inverse Rayleigh distribution and has the potential to model the recovery times of Covid-19 patients.The new distribution is obtained using the distributions of the first two lower record statistics of the inverse Rayleigh distribution. We discuss some statistical inferences and mathematical properties of the suggested distribution. We examine some characteristics of the proposed distribution such as density shape, hazard function,moments, moment generating function, incomplete moments,Rényi entropy, order statistics, stochastic ordering. We consider five estimation methods such as maximum likelihood, least squares, weighted least squares, Anderson-Darling, Cramér-von Mises for the point estimation of the proposed distribution. Then, a comprehensive Monte Carlo simulation study is carried out to assess the risk behavior of the examined estimators. We provide two real data applications to illustrate the fitting ability of the proposed model, and compare its fit with competitor ones. Unlike many previously proposed distributions, the introduced distribution in this paper has modeled the recovery times of Covid-19 patients.