An overview of model-robust regression

The proper combination of parametric and nonparametric regression procedures can improve upon the shortcomings of each when used individually. Considered is the situation where the researcher has an idea of which parametric model should explain the behavior of the data, but this model is not adequate throughout the entire range of the data. An extension of partial linear regression and two other methods of model-robust regression are developed and compared in this context. The model-robust procedures each involve the proportional mixing of a parametric fit to the data and a nonparametric fit to either the data or residuals. The emphasis of this work is on fitting in the small-sample situation, where nonparametric regression alone has well-known inadequacies. Performance is based on bias and variance considerations, and theoretical mean squared error formulas are developed for each procedure. An example is given that uses generated data from an underlying model with defined misspecification to provide graphical comparisons of the fits and to show the theoretical benefits of the model-robust procedures. Simulation results are presented which establish the accuracy of the theoretical formulas and illustrate the potential benefits of the model-robust procedures. Simulations are also used to illustrate the advantageous properties of a data-driven selector developed in this work for choosing the smoothing and mixing parameters. It is seen that the model-robust procedures (the final proposed method, in particular) give much improved fits over the individual parametric and nonparametric fits.     

Rate-optimal nonparametric estimation in classical and Berkson errors-in-variables problems

We consider nonparametric estimation of a regression curve when the data are observed with Berkson errors or with a mixture of classical and Berkson errors. In this context, other existing nonparametric procedures can either estimate the regression curve consistently on a very small interval or require complicated inversion of an estimator of the Fourier transform of a nonparametric regression estimator. We introduce a new estimation procedure which is simpler to implement, and study its asymptotic properties. We derive convergence rates which are faster than those previously obtained in the literature, and we prove that these rates are optimal. We suggest a data-driven bandwidth selector and apply our method to some simulated examples.     

Robust plug-in bandwidth estimators in nonparametric regression

In this paper, we propose a robust bandwidth selection method for local M-estimates used in nonparametric regression. We study the asymptotic behavior of the resulting estimates. We use the results of a Monte Carlo study to compare the performance of various competitors for moderate samples sizes. It appears that the robust plug-in bandwidth selector we propose compares favorably to its competitors, despite the need to select a pilot bandwidth. The Monte Carlo study shows that the robust plug-in bandwidth selector is very stable and relatively insensitive to the choice of the pilot.     

On completely data-driven bandwidth selection for single-index models

The class of single-index models (SIMs) has become an important tool for nonparametric regression analysis. As with any other nonparametric regression models, the selection of bandwidth plays an important role in the inferences of the SIMs. However, most results in the literature either take the bandwidths as externally given, or require unpractical assumptions or very restrictive conditions for data-driven bandwidths. We examine the asymptotic properties of a popular bandwidth selection method based on cross-validation that is completely data-driven, under much weaker conditions than those assumed in the literature. And we show that the same bandwidth that is optimal for estimating the index vector, can be used for nearly optimal error variance estimation through the method of varying cross-validation. A simulation study is presented to demonstrate the finite sample performance of the proposed procedures, based on which we recommend a simple 2-step procedure for bandwidth selection, index vector estimation, as well as error variance estimation.     

Automatic bandwidth selection for modified m-smoother ∗

Traditional parametric and nonparametric regression techniques encounter serious over smoothing problems when jump point discontinuities exist in the underlying mean function. Recently, Chu, Glad, Godtliebsen and Marron (1998) developed a method using a modified M-smoothing technique to preserve jumps and spikes while producing a smooth estimate of the mean function. The performance of Chu etal.'s (1998) method is quite sensitive to the choice of the required bandwidths g and h. Furthermore, it is not obvious how to extend certain commonly used automatic bandwidth selection procedures when jumps and spikes are present. In this paper we propose a rule of thumb method of choosing the smoothing parameters based on asymptotic optimal bandwidth formulas and robust estimates of unknown quantities. We also evaluate the proposed bandwidth selection method via a small simulation study.     

On the nonparametric smooth estimation of the reversed Hazard Rate function

The Reversed Hazard Rate (RHR) function is an important measure as a tool in the analysis of the reliability of both natural and man-made systems. In this paper, we present several new estimators of the RHR function using nonparametric techniques. These estimators are obtained by incorporating different binning techniques with fixed design local polynomial regression. We show that these estimators are asymptotically unbiased and consistent and, to determine the bandwidth, we propose two simple yet efficient plug-in bandwidth selection methods for even and odd order local polynomial estimators. Simulated and real life data are subsequently used to evaluate the performances of these estimators.     

Discrete Nonparametric Kernel and Parametric Methods for the Modeling of Pavement Deterioration

This article is concerned with one discrete nonparametric kernel and two parametric regression approaches for providing the evolution law of pavement deterioration. The first parametric approach is a survival data analysis method; and the second is a nonlinear mixed-effects model. The nonparametric approach consists of a regression estimator using the discrete associated kernels. Some asymptotic properties of the discrete nonparametric kernel estimator are shown as, in particular, its almost sure consistency. Moreover, two data-driven bandwidth selection methods are also given, with a new theoretical explicit expression of optimal bandwidth provided for this nonparametric estimator. A comparative simulation study is realized with an application of bootstrap methods to a measure of statistical accuracy.     

Plug-in bandwidth choice for estimation of nonparametric part in partial linear regression models with strong mixing errors

Consider a regression model where the regression function is the sum of a linear and a nonparametric component. Assuming that the errors of the model follow a stationary strong mixing process with mean zero, the problem of bandwidth selection for a kernel estimator of the nonparametric component is addressed here. We obtain an asymptotic expression for an optimal band-width and we propose to use a plug-in methodology in order to estimate this bandwidth through preliminary estimates of the unknown quantities. Asymptotic optimality for the plug-in bandwidth is established.     

New aspects of Bregman divergence in regression and classification with parametric and nonparametric estimation

In statistical learning, regression and classification concern different types of the output variables, and the predictive accuracy is quantified by different loss functions. This article explores new aspects of Bregman divergence (BD), a notion which unifies nearly all of the commonly used loss functions in regression and classification. The authors investigate the duality between BD and its generating function. They further establish, under the framework of BD, asymptotic consistency and normality of parametric and nonparametric regression estimators, derive the lower bound of their asymptotic covariance matrices, and demonstrate the role that parametric and nonparametric regression estimation play in the performance of classification procedures and related machine learning techniques. These theoretical results and new numerical evidence show that the choice of loss function affects estimation procedures, whereas has an asymptotically relatively negligible impact on classification performance. Applications of BD to statistical model building and selection with non‐Gaussian responses are also illustrated. The Canadian Journal of Statistics 37: 119‐139; 2009 © 2009 Statistical Society of Canada     

Bandwidth selection for local linear regression smoothers

Summary. The paper presents a general strategy for selecting the bandwidth of nonparametric regression estimators and specializes it to local linear regression smoothers. The procedure requires the sample to be divided into a training sample and a testing sample. Using the training sample we first compute a family of regression smoothers indexed by their bandwidths. Next we select the bandwidth by minimizing the empirical quadratic prediction error on the testing sample. The resulting bandwidth satisfies a finite sample oracle inequality which holds for all bounded regression functions. This permits asymptotically optimal estimation for nearly any regression function. The practical performance of the method is illustrated by a simulation study which shows good finite sample behaviour of our method compared with other bandwidth selection procedures.     

A weighted polynomial regression method for local fitting of spatial data

Typically, parametric approaches to spatial problems require restrictive assumptions. On the other hand, in a wide variety of practical situations nonparametric bivariate smoothing techniques has been shown to be successfully employable for estimating small or large scale regularity factors, or even the signal content of spatial data taken as a whole.We propose a weighted local polynomial regression smoother suitable for fitting of spatial data. To account for spatial variability, we both insert a spatial contiguity index in the standard formulation, and construct a spatial-adaptive bandwidth selection rule. Our bandwidth selector depends on the Gearys local indicator of spatial association. As illustrative example, we provide a brief Monte Carlo study case on equally spaced data, the performances of our smoother and the standard polynomial regression procedure are compared.This note, though it is the result of a close collaboration, was specifically elaborated as follows: paragraphs 1 and 2 by T. Sclocco and the remainder by M. Di Marzio. The authors are grateful to the referees for constructive comments and suggestions.     

Asymptotic distribution of bandwidth selectors in kernel regression estimation

Gasser, Kneip and Köhler (1991) proposed a fast and flexible procedure for automatic bandwidth selection in kernel regression estimation. This article describes this method and additionally derives the joint asymptotic normal distribution of this bandwidth selector with the realizationwise optimal bandwidth.     

Data-Based Choice of the Number of Pilot Stages for Plug-in Bandwidth Selection

The choice of the bandwidth is a crucial issue for kernel density estimation. Among all the data-dependent methods for choosing the bandwidth, the direct plug-in method has shown a particularly good performance in practice. This procedure is based on estimating an asymptotic approximation of the optimal bandwidth, using two “pilot” kernel estimation stages. Although two pilot stages seem to be enough for most densities, for a long time the problem of how to choose an appropriate number of stages has remained open. Here we propose an automatic (i.e., data-based) method for choosing the number of stages to be employed in the plug-in bandwidth selector. Asymptotic properties of the method are presented and an extensive simulation study is carried out to compare its small-sample performance with that of the most recommended bandwidth selectors in the literature.     

Using difference-based methods for inference in nonparametric regression with time series errors

Summary. We show that difference-based methods can be used to construct simple and explicit estimators of error covariance and autoregressive parameters in nonparametric regression with time series errors. When the error process is Gaussian our estimators are efficient, but they are available well beyond the Gaussian case. As an illustration of their usefulness we show that difference-based estimators can be used to produce a simplified version of time series cross-validation. This new approach produces a bandwidth selector that is equivalent, to both first and second orders, to that given by the full time series cross-validation algorithm. Other applications of difference-based methods are to variance estimation and construction of confidence bands in nonparametric regression.     

Local maximum likelihood estimation and inference

Local maximum likelihood estimation is a nonparametric counterpart of the widely used parametric maximum likelihood technique. It extends the scope of the parametric maximum likelihood method to a much wider class of parametric spaces. Associated with this nonparametric estimation scheme is the issue of bandwidth selection and bias and variance assessment. This paper provides a unified approach to selecting a bandwidth and constructing confidence intervals in local maximum likelihood estimation. The approach is then applied to least squares nonparametric regression and to nonparametric logistic regression. Our experiences in these two settings show that the general idea outlined here is powerful and encouraging.     

Estimation for semiparametric varying coefficient models with different smoothing variables under random right censoring

In this paper we study a semiparametric varying coefficient model when the response is subject to random right censoring. The model gives an easy interpretation due to its direct connectivity to the classical linear model and is very flexible since nonparametric functions which accommodates various nonlinear interaction effects between covariates are admitted in the model. We propose estimators for this model using mean-preserving transformation and establish their asymptotic properties. The estimation procedure is based on the profiling and the smooth backfitting techniques. A simulation study is presented to show the reliability of the proposed estimators and an automatic bandwidth selector is given in a data-driven way.     

Empirical likelihood in a regression model with noised variables

This paper is concerned with a recently developed regression model with noised variables in which the means of the response and some covariable components are nonparametric functions of an auxiliary variable. Previous results have shown that the de-noised estimators of the parameters of interest are asymptotically normal when undersmoothing is applied. But undersmoothing causes difficulties in bandwidth selection. To avoid this problem, we propose an empirical log-likelihood ratio for the regression coefficients and derive a nonparametric version of Wilk's theorem. The confidence region based on the empirical likelihood has three advantages compared with those based on asymptotic normality: (1) It does not have the predetermined symmetry, which enables it to better reflect the true shape of the underlying distribution; (2) it does not involve any asymptotic covariance matrix estimation and hence is robust against the heteroscedasticity; and (3) it avoids undersmoothing the regressor functions so that optimal bandwidth can be used. A small simulation is conducted to compare the finite sample performances of these two methods. An example of application on a set of advertising data is also illustrated.     

Finite Mixture of Generalized Semiparametric Models: Variable Selection via Penalized Estimation

Selection of the important variables is one of the most important model selection problems in statistical applications. In this article, we address variable selection in finite mixture of generalized semiparametric models. To overcome computational burden, we introduce a class of variable selection procedures for finite mixture of generalized semiparametric models using penalized approach for variable selection. Estimation of nonparametric component will be done via multivariate kernel regression. It is shown that the new method is consistent for variable selection and the performance of proposed method will be assessed via simulation.     

Quantile regression for robust estimation and variable selection in partially linear varying-coefficient models

In this paper, we develop a new estimation procedure based on quantile regression for semiparametric partially linear varying-coefficient models. The proposed estimation approach is empirically shown to be much more efficient than the popular least squares estimation method for non-normal error distributions, and almost not lose any efficiency for normal errors. Asymptotic normalities of the proposed estimators for both the parametric and nonparametric parts are established. To achieve sparsity when there exist irrelevant variables in the model, two variable selection procedures based on adaptive penalty are developed to select important parametric covariates as well as significant nonparametric functions. Moreover, both these two variable selection procedures are demonstrated to enjoy the oracle property under some regularity conditions. Some Monte Carlo simulations are conducted to assess the finite sample performance of the proposed estimators, and a real-data example is used to illustrate the application of the proposed methods.     

NEW EFFICIENT ESTIMATION AND VARIABLE SELECTION METHODS FOR SEMIPARAMETRIC VARYING-COEFFICIENT PARTIALLY LINEAR MODELS

The complexity of semiparametric models poses new challenges to statistical inference and model selection that frequently arise from real applications. In this work, we propose new estimation and variable selection procedures for the semiparametric varying-coefficient partially linear model. We first study quantile regression estimates for the nonparametric varying-coefficient functions and the parametric regression coefficients. To achieve nice efficiency properties, we further develop a semiparametric composite quantile regression procedure. We establish the asymptotic normality of proposed estimators for both the parametric and nonparametric parts and show that the estimators achieve the best convergence rate. Moreover, we show that the proposed method is much more efficient than the least-squares-based method for many non-normal errors and that it only loses a small amount of efficiency for normal errors. In addition, it is shown that the loss in efficiency is at most 11.1% for estimating varying coefficient functions and is no greater than 13.6% for estimating parametric components. To achieve sparsity with high-dimensional covariates, we propose adaptive penalization methods for variable selection in the semiparametric varying-coefficient partially linear model and prove that the methods possess the oracle property. Extensive Monte Carlo simulation studies are conducted to examine the finite-sample performance of the proposed procedures. Finally, we apply the new methods to analyze the plasma beta-carotene level data.