A note on nonparametric density deconvolution by weighted kernel estimators

다국어 초록 (Multilingual Abstract)

Recently Hazelton and Turlach (2009) proposed a weighted kernel density estimator for the deconvolution problem. In the case of Gaussian kernels and measurement error, they argued that the weighted kernel density estimator is a competitive estimator over the classical deconvolution kernel estimator. In this paper we consider weighted kernel density estimators when sample observations are contaminated by double exponentially distributed errors. The performance of the weighted kernel density estimators is compared over the classical deconvolution kernel estimator and the kernel density estimator based on the support vector regression method by means of a simulation study.
The weighted density estimator with the Gaussian kernel shows numerical instability in practical implementation of optimization function. However the weighted density estimates with the double exponential kernel has very similar patterns to the classical kernel density estimates in the simulations, but the shape is less satisfactory than the classical kernel density estimator with the Gaussian kernel.

참고문헌 (Reference)

1 Louis, T. A., "Using empirical Bayes methods in biopharmaceutical research" 10 : 811-827, 1991

2 Mukherjee, S, "Support vector method for multivariate density estimation" 659-665, 1999

3 Gunn, S. R., "Support vector machines for classi cation and regression" University of Southampton 1998

4 Carroll, R. J., "Optimal rates of convergence for deconvoluting a density" 83 : 1184-1886, 1988

5 Fan, J., "On the optimal rates of convergence for nonparametric deconvolution problem" 19 : 1257-1272, 1991

6 Hazelton, M. L, "Nonparametric density deconvolution by weighted kernel esti-mators" 19 : 217-228, 2009

7 Eggermont, P., "Nonlinearly smoothed EM density estimation with augmented smoothing parameter selection for nonparametric deconvolution problems" 92 : 1451-1458, 1997

8 Carroll, R. J., "Measurement error in nonlinear models" Chapman and Hall 1995

9 Dilaigle, A, "Frequent problems in calculating integrals and optimizing objective functions: A case study in density estimation" 17 : 349-355, 2007

10 Hall, P., "Discrete-transform approach to deconvolution problems" 92 : 135-148, 2005

11 Mendelsohn, J., "Deconvolution of micro uorometric histograms with B splines" 77 : 748-753, 1982

12 Wang, X., "Deconvolution estimation in measurement error models: The R package decon" 39 (39): 2011

13 Stefanski, L., "Deconvoluting kernel density estimators" 21 : 169-184, 1990

14 Koo, J.-Y., "B-spline deconvolution based on the EM algorithm" 54 : 275-288, 1996

15 Pensky, M., "Adaptive wavelet estimator for nonparametric density deconvolutoin" 27 : 2033-2053, 1999

16 Lee, S., "A note on support vector density estimation for the deconvolution problem" 37 : 328-336, 2008

17 이성호, "A Support Vector Method for the Deconvolution Problem" 한국통계학회 17 (17): 451-457, 2010

18 Liu, M. C, "A Consistent nonparametric density estimator for the deconvolution problem" 17 : 427-438, 1989

19 이성호, "A Note on Deconvolution Estimators when Measurement Errors are Normal" 한국통계학회 19 (19): 517-526, 2012