Fast Algorithm for Choosing Kernel Function Blur Coefficients in a Nonparametric Probability Density Estimate
- 31 Downloads
A fast algorithm for choosing the blurring coefficients of kernel functions for a nonparametric probability density estimate is proposed, and its properties are investigated. The technique of interval estimation of the standard deviation of the nonparametric statistics under consideration is considered.
Keywordsnonparametric estimation of probability density choice of blurring coefficients estimation of mean square deviation of probability density estimation
This study was supported by the Russian Foundation for Basic Research (Grant No. 18-01-00251).
- 1.A. V. Lapko and V. A. Lapko, Multilevel Non-Parametric Information Processing Systems, Sib-GAU, Krasnoyarsk (2013).Google Scholar
- 2.M. Rudemo, “Empirical choice of histogram and kernel density estimators,” Scand. J. Statist., No. 9, 65–78 (1982).Google Scholar
- 7.A. V. Lapko and V. A. Lapko, “Analysis of optimization methods for nonparametric estimation of probability density with respect to the blur factor of kernel functions,” Izmer. Tekhn., No. 6, 3–8 (2017).Google Scholar
- 12.V. C. Raykar and R. Duraiswami, “Fast optimal bandwidth selection for kernel density estimation,” 6th SIAM Int. Conf. on Data Mining (2006), pp. 524–528.Google Scholar
- 14.L. Devroi and L. Dierfi , Nonparametric Density Estimation (L 1 -approach), Mir, Moscow (1988).Google Scholar
- 16.A. V. Lapko and V. A. Lapko, “Regression estimate of the multidimensional probability density and its properties,” Optoelectr., Instrum. Data Proces., 5, No. 2, 148–153 (2014).Google Scholar