Local Polynomial Smoothing Research Articles

Local quasi-likelihood with a parametric guide

Generalized linear models and quasi-likelihood method extend the ordinary regression models to accommodate more general conditional distributions of the response. Nonparametric methods need no explicit parametric specification and the resulting model is completely determined by the data themselves. However nonparametric estimation schemes generally have a slower convergence rate such as the local polynomial smoothing estimation of nonparametric generalized linear models studied in Fan, Heckman and Wand (1995). In this work, we propose two parametrically guided nonparametric estimation schemes by incorporating prior shape information on the link transformation of the response variable's conditional mean in terms of the predictor variable. Asymptotic results and numerical simulations demonstrate the improvement of our new estimation schemes over the original nonparametric counterpart.

GEE Modeling with Longitudinal Binary Data: Goodness-of-Fit Assessment via Local Polynomial Smoothing

Analysis of longitudinal binary data is often accomplished by using GEE methodology to estimate the marginal model parameters. Most of current goodness-of-fit tests for GEE models have been studied in parametric situations. In this article, we consider to develop an alternative assessment for GEE models utilizing nonparametric technique. The proposed test avoided the explosion of a large number of additional parameters and dependence on partition of covariate space. Even though exact expectation and variance of the proposed test statistic are analytically and computationally infeasible, approximated values based on bootstrap data are employed. The asymptotic distribution of the proposed test statistic in terms of a scaled chi-squared distribution, and comparison of the proposed test and the current methods with respect to power are discussed by simulation studies. In addition, the testing procedure is illustrated by a medical study from Koch et al. [12].

Efficient Estimation of Three-Dimensional Curves and their Derivatives by Free-Knot Regression Splines, Applied to the Analysis of Inner Carotid Artery Centrelines

Summary We deal with the problem of efficiently estimating a three-dimensional curve and its derivatives, starting from a discrete and noisy observation of the curve. This problem is now arising in many applicative contexts, thanks to the advent of devices that provide three-dimensional images and measures, such as three-dimensional scanners in medical diagnostics. Our research, in particular, stems from the need for accurate estimation of the curvature of an artery, from image reconstructions of three-dimensional angiographies. This need has emerged within the AneuRisk project, a scientific endeavour which aims to investigate the role of vessel morphology, blood fluid dynamics and biomechanical properties of the vascular wall, on the pathogenesis of cerebral aneurysms. We develop a regression technique that exploits free-knot splines in a novel setting, to estimate three-dimensional curves and their derivatives. We thoroughly compare this technique with a classical regression method, local polynomial smoothing, showing that three-dimensional free-knot regression splines yield more accurate and efficient estimates.

A Geometric Approach to Maximum Likelihood Estimation of the Functional Principal Components From Sparse Longitudinal Data

In this article, we consider the problem of estimating the eigenvalues and eigenfunctions of the covariance kernel (i.e., the functional principal components) from sparse and irregularly observed longitudinal data. We exploit the smoothness of the eigenfunctions to reduce dimensionality by restricting them to a lower dimensional space of smooth functions. We then approach this problem through a restricted maximum likelihood method. The estimation scheme is based on a Newton–Raphson procedure on the Stiefel manifold using the fact that the basis coefficient matrix for representing the eigenfunctions has orthonormal columns. We also address the selection of the number of basis functions, as well as that of the dimension of the covariance kernel by a second-order approximation to the leave-one-curve-out cross-validation score that is computationally very efficient. The effectiveness of our procedure is demonstrated by simulation studies and an application to a CD4+ counts dataset. In the simulation studies, our method performs well on both estimation and model selection. It also outperforms two existing approaches: one based on a local polynomial smoothing, and another using an EM algorithm. Supplementary materials including technical details, the R package fpca, and data analyzed by this article are available online.

Estimating the error distribution function in nonparametric regression with multivariate covariates

We consider nonparametric regression models with multivariate covariates and estimate the regression curve by an undersmoothed local polynomial smoother. The resulting residual-based empirical distribution function is shown to differ from the error-based empirical distribution function by the density times the average of the errors, up to a uniformly negligible remainder term. This result implies a functional central limit theorem for the residual-based empirical distribution function.

Real time estimation in local polynomial regression, with application to trend-cycle analysis

The paper focuses on the adaptation of local polynomial filters at the end of the sample period. We show that for real time estimation of signals (i.e., exactly at the boundary of the time support) we cannot rely on the automatic adaptation of the local polynomial smoothers, since the direct real time filter turns out to be strongly localized, and thereby yields extremely volatile estimates. As an alternative, we evaluate a general family of asymmetric filters that minimizes the mean square revision error subject to polynomial reproduction constraints; in the case of the Henderson filter it nests the well-known Musgrave’s surrogate filters. The class of filters depends on unknown features of the series such as the slope and the curvature of the underlying signal, which can be estimated from the data. Several empirical examples illustrate the effectiveness of our proposal.

Shrinkage Estimation of the Varying Coefficient Model

The varying coefficient model is a useful extension of the linear regression model. Nevertheless, how to conduct variable selection for the varying coefficient model in a computationally efficient manner is poorly understood. To solve the problem, we propose here a novel method, which combines the ideas of the local polynomial smoothing and the Least Absolute Shrinkage and Selection Operator (LASSO). The new method can do nonparametric estimation and variable selection simultaneously. With a local constant estimator and the adaptive LASSO penalty, the new method can identify the true model consistently, and that the resulting estimator can be as efficient as the oracle estimator. Numerical studies clearly confirm our theories. Extension to other shrinkage methods (e.g., the SCAD, i.e., the Smoothly Clipped Absolute Deviation.) and other smoothing methods is straightforward.

Real Time Estimation in Local Polynomial Regression, with Application to Trend-Cycle Analysis

The paper focuses on the adaptation of local polynomial filters at the end of the sample period. We show that for real time estimation of signals (i.e. exactly at the boundary of the time support) we cannot rely on the automatic adaptation of the local polynomial smoothers, since the direct real time filter turns out to be strongly localised, and thereby yields extremely volatile estimates. As an alternative we evaluate a general family of asymmetric filters that minimises the mean square revision error subject to polynomial reproduction constraints; in the case of the Henderson filter it nests the well known Musgrave's surrogate filters. The class of filters depends on unknown features of the series such as the slope and the curvature of the underlying signal, which can be estimated from the data. Several empirical examples illustrate the effectiveness of our proposal. We also discuss the merits of using a nearest neighbour bandwidth as opposed to a fixed bandwidth for improving the quality of the approximation.

Local Polynomial Estimation for Sensitivity Analysis on Models With Correlated Inputs

Sensitivity indexes when the inputs of a model are not independent are derived from local polynomial techniques. Two original estimators based on local polynomial smoothers are proposed. Both have good theoretical properties, which are illustrated through analytical examples. Comparison with the Bayesian approach developed by Oakley and O’Hagan (2004) is also performed. The two proposed estimators are used to carry out a sensitivity analysis on two real case models with correlated parameters.

Shrinkage Estimation of the Varying Coefficient Model

The varying coefficient model is a useful extension of the linear regression model. Nevertheless, how to conduct variable selection for the varying coefficient model in a computationally efficient manner is poorly understood. To solve the problem, we propose here a novel method, which combines the ideas of the local polynomial smoothing (Fan and Zhang, 1999) and the shrinkage estimation (Tibshirani, 1996, LASSO). The new method can do nonparametric estimation and variable selection simultaneously. With a local constant estimator and the adaptive LASSO penalty, the new method can identify the true model consistently, and that the resulting estimator can be as efficient as the oracle estimator (Fan and Li, 2001). Numerical studies clearly confirm our theories. Extension to other shrinkage methods (e.g., the SCAD) and other smoothing methods (Zhang and Lin, 2003) is straightforward.

Generalized likelihood ratio tests for the structure of semiparametric additive models

AbstractSemiparametric additive models (SAMs) are very useful in multivariate nonparametric regression. In this paper, the authors study nonparametric testing problems for the nonparametric components of SAMs. Using the backfitting algorithm and the local polynomial smoothing technique, they extend to SAMs the generalized likelihood ratio tests of Fan &Jiang (2005). The authors show that the proposed tests possess the Wilks‐type property and that they can detect alternatives nearing the null hypothesis with a rate arbitrarily close to root‐n while error distributions are unspecified. They report simulations which demonstrate the Wilks phenomenon and the powers of their tests. They illustrate the performance of their approach by simulation and using the Boston housing data set.

On extracting information implied in options

Options are financial instruments with a payoff depending on future states of the underlying asset. Therefore option markets contain information about expectations of the market participants about market conditions, e.g. current uncertainty on the market and corresponding risk. A standard measure of risk calculated from plain vanilla options is the implied volatility (IV). IV can be understood as an estimate of the volatility of returns in future period. Another concept based on the option markets is the state-price density (SPD) that is a density of the future states of the underlying asset. From raw data we can recover the IV function by nonparametric smoothing methods. Smoothed IV estimated by standard techniques may lead to a non-positive SPD which violates no arbitrage criteria. In this paper, we combine the IV smoothing with SPD estimation in order to correct these problems. We propose to use the local polynomial smoothing technique. The elegance of this approach is that it yields all quantities needed to calculate the corresponding SPD. Our approach operates only on the IVs--a major improvement comparing to the earlier multi-step approaches moving through the Black---Scholes formula from the prices to IVs and vice-versa.

Rates of strong uniform consistency for local least squares kernel regression estimators

We establish exact rates of strong uniform consistency for the multivariate Nadaraya–Watson kernel estimator of the regression function and its derivatives. As a special case, we treat the local linear estimator of the regression and the local polynomial smoothers of derivatives of the regression in the more convenient univariate setting. Our methods of proofs are based upon modern empirical process theory in the spirit of the results of Einmahl and Mason [2000. An empirical process approach to the uniform consistency of kernel-type function estimators. J. Theoret. Probab. 13.1, 1–37.] and Deheuvels and Mason [2004. General asymptotic confidence bands based on kernel-type function estimators. Statist. Infer. Stochastic Process. 7.3, pp. 225–277] relative to uniform deviations of nonparametric kernel estimators.

An empirical study of a test for polynomial relationships in randomly right censored regression models

In this paper, a test statistic is constructed to test polynomial relationships in randomly right censored regression models based on the local polynomial smoothing technique. Two bootstrap procedures, namely the residual-based bootstrap and the naive bootstrap procedures, are suggested to derive the p -value of the test. Some simulations are conducted to empirically assess the performance of the two bootstrap procedures. The results demonstrate that the residual-based bootstrap performs much better than the naive bootstrap and the test method with the residual-based bootstrap to derive the p -value works satisfactorily. Although the limiting distribution of the test statistic and the consistency of the bootstrap approximations remain to be investigated, simulation results indicate that the proposed test method may be of some practical use. As a real example, the proposed test is applied to the Stanford heart transplant data.

Nonparametric Model Calibration Estimation in Survey Sampling

Calibration is commonly used in survey sampling to include auxiliary information at the estimation stage of a population parameter. Calibrating the observation weights on population means (totals) of a set of auxiliary variables implies building weights that when applied to the auxiliaries give exactly their population mean (total). Implicitly, calibration techniques rely on a linear relation between the survey variable and the auxiliary variables. However, when auxiliary information is available for all units in the population, more complex modeling can be handled by means of model calibration; auxiliary variables are used to obtain fitted values of the survey variable for all units in the population, and estimation weights are sought to satisfy calibration constraints on the fitted values population mean, rather than on the auxiliary variables one. In this work we extend model calibration considering more general superpopulation models and use nonparametric methods to obtain the fitted values on which to calibrate. More precisely, we adopt neural network learning and local polynomial smoothing to estimate the functional relationship between the survey variable and the auxiliary variables. Under suitable regularity conditions, the proposed estimators are proven to be design consistent. The moments of the asymptotic distribution are also derived, and a consistent estimator of the variance of each distribution is then proposed. The performance of the proposed estimators for finite-size samples is investigated by means of simulation studies. An application to the assessment of the ecological conditions of streams in the mid-Atlantic highlands in the United States is also carried out.

Approximate and Asymptotic Distributions of Chi-Squared–Type Mixtures With Applications

In this article we study how to approximate a random variable T of general chi-squared–type mixtures by a random variable of the form via matching the first three cumulants. The approximation error bounds for the density functions of the chi-squared approximation and the normal approximation are established. Applications of the results to some nonparametric goodness-of-fit tests, including those tests based on orthogonal series, smoothing splines, and local polynomial smoothers, are investigated. Two simulation studies are conducted to compare the chi-squared approximation and the normal approximation numerically. The chi-squared approximation is illustrated using a real data example for polynomial goodness-of-fit tests.

Local polynomial regression analysis of clustered data

This paper proposes a classical weighted least squares type of local polynomial smoothing for the analysis of clustered data, with the key idea of using generalised inverses of correlation matrices. The estimator has a simple closed-form expression. Simplicity is achieved also for nonparametric generalised linear models with arbitrary link function via a transformation. Our approach can be characterised by ‘local observations with local variances’, which yields intuitively correct results in the sense that correct/incorrect specification of within-cluster correlation has respective positive/negative effects. The approach is a natural extension of classical local polynomial smoothing. Consequently, existing theory can be largely carried over and important issues such as bandwidth selection can be tackled in the classical fashion. Moreover, the approach can handle various types of covariate, such as cluster-level, subject-level or partially cluster-level. Numerical studies support the theoretical results. The method is illustrated with a real example on luteinising hormone levels in cows.

Local smoothing with robustness against outlying predictors

AbstractOutlying pollutant concentration data are frequently observed in time series studies conducted to investigate the effects of atmospheric pollution on mortality/morbidity. These outliers may severely affect the estimation procedures and even generate unexpected results like a protective effect of pollution. Although robust methods have been proposed to downweight the effect of outliers in the response variable distribution, little has been done to handle outlying explanatory variable values. We consider a robust local polynomial smoothing technique which may be useful for such purposes. It is based on downweighting points with a small design density and may also be used as a diagnostic tool to identify outliers. Using data from a study conducted in São Paulo, Brazil, we show how an unexpected form of the relative risk curve of mortality attributable to pollution by SO2 obtained via nonrobust methods may be completely reversed when the proposed technique is employed. Copyright © 2004 John Wiley & Sons, Ltd.

A power comparison between nonparametric regression tests

In this paper, we consider three major types of nonparametric regression tests that are based on kernel and local polynomial smoothing techniques. Their asymptotic power comparisons are established systematically under the fixed and contiguous alternatives, and are also illustrated through nonasymptotic investigations and finite-sample simulation studies.

Assessing the equivalence of nonparametric regression tests based on spline and local polynomial smoothers

It is widely known that, in a certain sense, a smoothing spline estimate of the regression function is asymptotically equivalent to a kernel regression estimate. However, little information has been available about the equivalence between nonparametric regression tests, based on the smoothing spline and local polynomial regression methods. To assess their relative behaviors and to facilitate illustration, we consider in this paper the “generalized likelihood ratio” (GLR) test statistic, constructed from each type of smoother. For the resulting test statistics, we first establish their equivalent forms of the asymptotic distributions, under the null hypothesis. After that, we derive their equivalent optimal rates of smoothing parameters for nonparametric testing. Furthermore, we evaluate their relative asymptotic efficiency and characterize its relation to the magnitude of the smoothing parameters. Finally, we illustrate the large-sample behaviors of the cubic smoothing spline and local linear regression methods, in the GLR tests, with results from small-scale simulations.