Abstract
This paper is a continuation of the authors' earlier work [1], where a version of the Tråvén's [2] Gaussian clustering neural network (being a recursive counterpart of the EM algorithm) has been investigated. A comparative simulation study of the Gaussian clustering algorithm [1], two versions of plug-in kernel estimators and a version of Friedman's projection pursuit algorithm are presented for two- and three-dimensional data. Simulations show that the projection pursuit algorithm is a good or a very good estimator, provided, however, that the number of projections is suitably chosen. Although practically confined to estimating normal mixtures, the simulations confirm general reliability of plug-in estimators, and show the same property of the Gaussian clustering algorithm. Indeed, the simulations confirm the earlier conjecture that this last estimator proivdes a way of effectively estimating arbitrary and highly structured continuous densities on Rd, at least for small d, either by using this estimator itself or, rather, by using it as a pilot estimator for a newly proposed plug-in estimator.
Similar content being viewed by others
References
Ćwik J, Koronacki J. Probability density estimation using a Gaussian clustering algorithm. Neural Comput & Applic 1996; 4: 149–160.
Tråvén HGC. A neural network approach to statistical pattern classification by semiparametric estimation of probability density functions. IEEE Trans. Neural Networks 1991; 2: 366–377.
Specht DF. Probabilistic Neural Networks. Neural Networks 1990; 3: 109–118.
Härdie W. Smoothing Techniques. Springer-Verlag, 1991.
Scott DW. Multivariate Density Estimation: Theory, practice, and visualization. Wiley, 1992.
Sheather SJ. The performance of six popular bandwidth selection methods on some real data sets. Comp Statist 1992; 7: 225–250.
Park BU, Turlach BA. Practical performance of several data driven bandwidth selectors. Comp Statist 1992; 7: 251–270.
Cao R, Cuevas A, Gonzales-Manteiga W. A comparative study of several smoothing methods in density estimation. Comp Statist Data Analysis 1994; 17: 153–176.
Ciesielski Z. Asymptotic nonparametric spline density estimation. Probab Math Statist 1991; 12: 1–24.
Wand MP, Marron JS, Ruppert D. Transformations in density estimation. J Am Statist Assoc 1991; 86: 343–353.
Ruppert D, Cline DBH. Bias reduction in kernel density estimation by smoothed empirical transformations. Ann Statist 1994; 22: 185–210.
Wand MP. Error analysis for general multivariate kernel estimators. J Nonpar Statist 1992; 2: 1–15.
Wand MP, Jones MC. Comparison of smoothing parametrizations in bivariate kernel density estimation. J Am Statist Assoc 1993; 88: 520–528.
Ciesielski Z. Asymptotic nonparametric spline density estimation in several variables. International Studies of Numerical Mathematics, Birkhauser, 1990; 94: 25–53.
O'Sullivan F, Pawitan Y. Multidimensional density estimation by tomography. J R Statist Soc B 1993; 55: 509–521.
Sain S, Baggerly K, Scott DW. Cross-validation of multivariate densities. J Am Statist Assoc 1994; 89: 807–817.
Wand MP, Jones MC. Multivariate plug-in bandwidth selection. Comp Statist 1994; 9: 97–116.
Titterington DM. Recursive parameter estimation using incomplete data. JR Statist Soc B 1984; 46: 257–267.
Sheather SJ, Jones MC. A reliable data-based bandwidth selection method for kernel density estimation. J Roy Statist Soc B 1991; 53: 683–690.
Wand MP, Jones MC. Comparison of smoothing parameterizations in bivariate kernel density estimation. Technical Report, Department of Statistics, Rice University, 1991.
Ćwik J and J Koronacki, A combined adaptivemixtures/plug-in estimator of multivariate probability densities. Comp Statist Data Analysis 1997 (to appear).
Friedman JH, Stuetzle W, Schroeder A. Projection pursuit density estimation. Am Statist Assoc 1984; 79: 599–608.
Friedman JH. Exploratory projection pursuit. J Am Statist Assoc 1987; 82: 249–266.
Huber PJ. Projection pursuit. Ann Statist 1985; 13: 435–475.
Priebe CE. Adaptive mixtures. J Am Statist Assoc 1994; 89: 796–806.
Silverman BW. Density Estimation for Statistics and Data Analysis. Chapman & Hall, 1986.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Ćwik, J., Koronacki, J. Multivariate density estimation: A comparative study. Neural Comput & Applic 6, 173–185 (1997). https://doi.org/10.1007/BF01413829
Issue Date:
DOI: https://doi.org/10.1007/BF01413829