Full bandwidth matrix selectors for gradient kernel density estimate

The most important factor in multivariate kernel density estimation is a choice of a bandwidth matrix. This choice is particularly important, because of its role in controlling both the amount and the direction of multivariate smoothing. Considerable attention has been paid to constrained parameterization of the bandwidth matrix such as a diagonal matrix or a pre-transformation of the data. A general multivariate kernel density derivative estimator has been investigated. Data-driven selectors of full bandwidth matrices for a density and its gradient are considered. The proposed method is based on an optimally balanced relation between the integrated variance and the integrated squared bias. The analysis of statistical properties shows the rationale of the proposed method. In order to compare this method with cross-validation and plug-in methods the relative rate of convergence is determined. The utility of the method is illustrated through a simulation study and real data applications.

[1]  D. W. Scott,et al.  Cross-Validation of Multivariate Densities , 1994 .

[2]  B. Silverman.  Density estimation for statistics and data analysis , 1986 .

[3]  Probal Chaudhuri,et al.  Significance in Scale Space for Bivariate Density Estimation , 2002 .

[4]  J. Magnus,et al.  Matrix Differential Calculus with Applications in Statistics and Econometrics (Revised Edition) , 1999 .

[5]  Kamila Vopatová,et al.  Visualization and Bandwidth Matrix Choice , 2012 .

[6]  M. Wand,et al.  ASYMPTOTICS FOR GENERAL MULTIVARIATE KERNEL DENSITY DERIVATIVE ESTIMATORS , 2011 .

[7]  Tarn Dunong.  Bandwidth selectors for multivariate kernel density estimation , 2005, Bulletin of the Australian Mathematical Society.

[8]  M. C. Jones,et al.  On a class of kernel density estimate bandwidth selectors , 1991 .

[9]  Jiří Zelinka,et al.  OPTIMAL CHOICE OF NONPARAMETRIC ESTIMATES OF A DENSITY AND OF ITS DERIVATIVES , 2002 .

[10]  J. Magnus,et al.  The Commutation Matrix: Some Properties and Applications , 1979 .

[11]  M. Hazelton,et al.  Cross‐validation Bandwidth Matrices for Multivariate Kernel Density Estimation , 2005 .

[12]  Jiří Zelinka,et al.  Contribution to the bandwidth choice for kernel density estimates , 2007, Comput. Stat..

[13]  M. Hazelton,et al.  Convergence rates for unconstrained bandwidth matrix selectors in multivariate kernal density estimation , 2005 .

[14]  A. Cuevas,et al.  A comparative study of several smoothing methods in density estimation , 1994 .

[15]  T. Duong,et al.  Multivariate plug-in bandwidth selection with unconstrained pilot bandwidth matrices , 2010 .

[16]  M. Wand,et al.  Multivariate plug-in bandwidth selection , 1994 .

[17]  J. Marron,et al.  SiZer for Exploration of Structures in Curves , 1999 .

[18]  James Stephen Marron,et al.  A simple root n bandwidth selector , 1991 .

[19]  J. Simonoff.  Smoothing Methods in Statistics , 1998 .

[20]  James Stephen Marron,et al.  Comparison of data-driven bandwith selectors , 1988 .

[21]  James Stephen Marron,et al.  Facts About the Gaussian Probability Density Function , 1995 .

[22]  M. Wand,et al.  Bandwidth choice for density derivatives , 1990 .

[23]  Jiří Zelinka,et al.  Bandwidth Choice for Kernel Density Estimates. , 2008 .

[24]  Inge Koch,et al.  Feature significance for multivariate kernel density estimation , 2008, Comput. Stat. Data Anal..

[25]  Charles C. Taylor,et al.  Bootstrap choice of the smoothing parameter in kernel density estimation , 1989 .

[26]  D. W. Scott,et al.  Biased and Unbiased Cross-Validation in Density Estimation , 1987 .

[27]  G. Terrell.  The Maximal Smoothing Principle in Density Estimation , 1990 .

[28]  D. W. Scott,et al.  Multivariate Density Estimation, Theory, Practice and Visualization , 1992 .

[29]  J. Marron,et al.  Transformations to reduce boundary bias in kernel density estimation , 1994 .

[30]  I. Horová,et al.  Kernel estimates of hazard functions for carcinoma data sets , 2006 .