Nonparametric kernel regression estimation near endpoints

When kernel regression is used to produce a smooth estimate of a curve over a finite interval, boundary problems detract from the global performance of the estimator. A new kernel is derived to reduce this boundary problem. A generalized jackknife combination of two unsatisfactory kernels produces the desired result. One motivation for adopting a jackknife combination is that they are simple to construct and evaluate. Furthermore, as in other settings, the bias reduction property need not cause an inordinate increase in variability. The convergence rate with the new boundary kernel is the same as for the non-boundary. To illustrate the general approach, a new second-order boundary kernel, which is continuously linked to the Epanechnikov (1969, Theory Probab. Appl. 14, 153–158) kernel, is produced. The asymptotic mean square efficiencies relative to smooth optimal kernels due to Gasser and Müller (1984, Scand. J. Statist. 11, 171–185), Müller (1991, Biometrika 78, 521–530) and Müller and Wang (1994, Biometrics 50, 61–76) indicate that the new kernel is also competitive in this sense.     

Nonparametric density estimation based on beta prime kernel

Abstract

In this work, we propose beta prime kernel estimator for estimation of a probability density functions defined with nonnegative support. For the proposed estimator, beta prime probability density function used as a kernel. It is free of boundary bias and nonnegative with a natural varying shape. We obtained the optimal rate of convergence for the mean squared error (MSE) and the mean integrated squared error (MISE). Also, we use adaptive Bayesian bandwidth selection method with Lindley approximation for heavy tailed distributions and compare its performance with the global least squares cross-validation bandwidth selection method. Simulation studies are performed to evaluate the average integrated squared error (ISE) of the proposed kernel estimator against some asymmetric competitors using Monte Carlo simulations. Moreover, real data sets are presented to illustrate the findings.     

Nonparametric kernel estimation of expected shortfall under negatively associated sequences

Abstract

In this article, Bahadur type expansions of a nonparametric kernel estimator for ES under NA sequences are given. The strong consistency and the uniformly asymptotic normality of the estimator are yielded from the Bahadur type expansions, while the convergence rates of the above asymptotic properties are also obtained. Moreover, the expectation, the variance and the mean squared error (MSE) of the estimator are given. Besides, the optimal bandwidth selection of this estimator is discussed. We point out that all above results are based on the NA sequences. Finally, we conduct numerical simulations and compare performances of some ES estimators.     

On nonparametric variogram estimation

A critical step for geostatistical prediction is estimation of variogram from the data. One of the popular methods estimating variogram is a smoothed version of classical nonparametric variogram estimator. In this paper we investigate its theoretical and empirical properties to provide useful information for using it. The main results are based on asymptotic theories (i.e., risk and central limit theorem) under nearly infill domain sampling. Simulation is also employed to make our points.     

Nonparametric estimation of value-at-risk

This paper develops a fully nonparametric method for estimating value-at-risk based on the adaptive volatility estimation and the nonparametric quantile estimation. The proposed method is simple, fast and easy to implement. We evaluated its numerical performance on the basis of Monte Carlo study for numerous models. We also provided an empirical application to KOrean Stock Price Index data, which turned out to be successful by backtesting.     

Nonparametric density deconvolution by weighted kernel estimators

Nonparametric density estimation in the presence of measurement error is considered. The usual kernel deconvolution estimator seeks to account for the contamination in the data by employing a modified kernel. In this paper a new approach based on a weighted kernel density estimator is proposed. Theoretical motivation is provided by the existence of a weight vector that perfectly counteracts the bias in density estimation without generating an excessive increase in variance. In practice a data driven method of weight selection is required. Our strategy is to minimize the discrepancy between a standard kernel estimate from the contaminated data on the one hand, and the convolution of the weighted deconvolution estimate with the measurement error density on the other hand. We consider a direct implementation of this approach, in which the weights are optimized subject to sum and non-negativity constraints, and a regularized version in which the objective function includes a ridge-type penalty. Numerical tests suggest that the weighted kernel estimation can lead to tangible improvements in performance over the usual kernel deconvolution estimator. Furthermore, weighted kernel estimates are free from the problem of negative estimation in the tails that can occur when using modified kernels. The weighted kernel approach generalizes to the case of multivariate deconvolution density estimation in a very straightforward manner.     

Nonparametric estimation of response coefficients

In this paper we propose a nonparametric kernel method of estimating response coefficients in the stochastic regressors model. The method is straightforward, and the estimator is easy to calculate. The asymptotic normality of the proposed estimator is established, and an illustrative example is presented.     

Nonparametric estimation of welfare changes

In this article, I propose a simple nonparametric method of estimating welfare measures with no structural assumptions on the demand system. The method is illustrated for both the single good and the multiple good case, using simulated data. Standard errors of the welfare changes are derived using standard bootstrap techniques. The compensating variation and deadweight loss resulting from a housing price increase is evaluated using Canadian mortgage data. The GAUSS coding, which is very short, is included.     

Nonparametric estimation of conditional expectation

Denote the integer lattice points in the N  -dimensional Euclidean space by Z^N${\mathbb{Z}}^N$ and assume that (_Xi,_Yi)$(X_{\mathbf{i}},Y_{\mathbf{i}})$, i∈Z^N$\mathbf{i}\in {\mathbb{Z}}^N$ is a mixing random field. Estimators of the conditional expectation r(x)=E[_Yi|_Xi=x]$r(\mathbf{x})=\mathrm{E}[Y_{\mathbf{i}}|X_{\mathbf{i}}=\mathbf{x}]$ by nearest neighbor methods are established and investigated. The main analytical result of this study is that, under general mixing assumptions, the estimators considered are asymptotically normal. Many difficulties arise since points in higher dimensional space N?2$N?2$ cannot be linearly ordered. Our result applies to many situations where parametric methods cannot be adopted with confidence.     

Nonparametric estimation of welfare changes

In this article, I propose a simple nonparametric method of estimating welfare measures with no structural assumptions on the demand system. The method is illustrated for both the single good and the multiple good case, using simulated data. Standard errors of the welfare changes are derived using standard bootstrap techniques. The compensating variation and deadweight loss resulting from a housing price increase is evaluated using Canadian mortgage data. The GAUSS coding, which is very short, is included.     

Nonparametric kernel estimation of the impact of tax policy on the demand for private health insurance in Australia

This paper is motivated by our attempt to answer a policy question: how is private health insurance take‐up in Australia affected by the income threshold at which the Medicare Levy Surcharge (MLS) kicks in? We propose a new difference deconvolution kernel estimator for the location and size of regression discontinuities. We also propose a bootstrapping procedure for estimating the confidence interval for the estimated discontinuity. Performance of the estimator is evaluated by Monte Carlo simulations before it is applied to estimating the effect of the income threshold of MLS on the take‐up of private health insurance in Australia, using contaminated data.     

Nonparametric models and their estimation

Summary Nonparametric models have become more and more popular over the last two decades. One reason for their popularity is software availability, which easily allows to fit smooth but otherwise unspecified functions to data. A benefit of the models is that the functional shape of a regression function is not prespecified in advance, but determined by the data. Clearly this allows for more insight which can be interpreted on a substance matter level. This paper gives an overview of available fitting routines, commonly called smoothing procedures. Moreover, a number of extensions to classical scatterplot smoothing are discussed, with examples supporting the advantages of the routines.     

Variable kernel density estimation

This paper considers the problem of selecting optimal bandwidths for variable (sample‐point adaptive) kernel density estimation. A data‐driven variable bandwidth selector is proposed, based on the idea of approximating the log‐bandwidth function by a cubic spline. This cubic spline is optimized with respect to a cross‐validation criterion. The proposed method can be interpreted as a selector for either integrated squared error (ISE) or mean integrated squared error (MISE) optimal bandwidths. This leads to reflection upon some of the differences between ISE and MISE as error criteria for variable kernel estimation. Results from simulation studies indicate that the proposed method outperforms a fixed kernel estimator (in terms of ISE) when the target density has a combination of sharp modes and regions of smooth undulation. Moreover, some detailed data analyses suggest that the gains in ISE may understate the improvements in visual appeal obtained using the proposed variable kernel estimator. These numerical studies also show that the proposed estimator outperforms existing variable kernel density estimators implemented using piecewise constant bandwidth functions.     

Nonparametric reliability estimation of semi-Markov processes

We define estimators of reliability and availability of such systems and show that they are uniformly strongly consistent and that each of them converges weakly to a normal random variable. The result of Baxter and Li (Scand. J. Statist. 21 (1994) 277) for an alternated renewal process appears as a particular case of our result. A method for constructing confidence intervals is also given. This is made possible by use of recent results of Ouhbi and Limnios (Appl. Stochast. Mod. Data Anal. 12 (4) (1996) 209; C. R. Acad. Sci. Paris 325 (1997) 921) establishing asymptotic properties of the estimators of the semi-Markov kernel and the Markov renewal matrix, respectively.     

Nonparametric estimation of extreme conditional quantiles

The estimation of extreme conditional quantiles is an important issue in different scientific disciplines. Up to now, the extreme value literature focused mainly on estimation procedures based on independent and identically distributed samples. Our contribution is a two-step procedure for estimating extreme conditional quantiles. In a first step nonextreme conditional quantiles are estimated nonparametrically using a local version of [Koenker, R. and Bassett, G. (1978). Regression quantiles. Econometrica, 46, 33–50.] regression quantile methodology. Next, these nonparametric quantile estimates are used as analogues of univariate order statistics in procedures for extreme quantile estimation. The performance of the method is evaluated for both heavy tailed distributions and distributions with a finite right endpoint using a small sample simulation study. A bootstrap procedure is developed to guide in the selection of an optimal local bandwidth. Finally the procedure is illustrated in two case studies.     

Nonparametric estimation of mean residual functions

Situations frequently arise in practice in which mean residual life (mrl) functions must be ordered. For example, in a clinical trial of three experiments, let e (1), e (2) and e (3) be the mrl functions, respectively, for the disease groups under the standard and experimental treatments, and for the disease-free group. The well-documented mrl functions e (1) and e (3) can be used to generate a better estimate for e (2) under the mrl restriction e (1) < or = e (2) < or = e (3). In this paper we propose nonparametric estimators of the mean residual life function where both upper and lower bounds are given. Small and large sample properties of the estimators are explored. Simulation study shows that the proposed estimators have uniformly smaller mean squared error compared to the unrestricted empirical mrl functions. The proposed estimators are illustrated using a real data set from a cancer clinical trial study.     

Nonparametric estimation of the limiting availability

The point availability of a repairable system is the probability that the system is operating at a specified time. As time increases, the point availability converges to a positive constant called the limiting availability. Baxter and Li (1994a) developed a technique for constructing nonparametric confidence intervals for the point availability. However, nonparametric estimators of the limiting availability have not previously been studied in the literature. In this paper, we consider two separate cases: (1) the data are complete and (2) the data are subject to right censorship. For each case, a nonparametric confidence interval for the limiting availability is derived. Applications and simulation studies are presented.deceased after the paper was accepted     

Nonparametric estimation of discrete hazard functions

A smoothing procedure for discrete time failure data is proposed which allows for the inclusion of covariates. This purely nonparametric method is based on discrete or continuous kernel smoothing techniques that gives a compromise between the data and smoothness. The method may be used as an exploratory tool to uncover the underlying structure or as an alternative to parametric methods when prediction is the primary objective. Confidence intervals are considered and alternative techniques of cross validation based choices of smoothing parameters are investigated.