Nonparametric Estimation of Heavy-Tailed Density by the Discrepancy Method
The nonparametric estimation of the probability density function (pdf) requires smoothing parameters like bandwidths of kernel estimates. We consider the so-called discrepancy method proposed in [13, 14, 21] as a data-driven smoothing tool and alternative to cross-validation. It is based on the von Mises-Smirnov’s (M-S) and the Kolmogorov–Smirnov’s (K-S) nonparametric statistics as measures in the space of distribution functions (cdfs). The unknown smoothing parameter is found as a solution of the discrepancy equation. On its left-hand side stands the measure between the empirical distribution function and the nonparametric estimate of the cdf. The latter is obtained as a corresponding integral of the pdf estimator. The right-hand side is equal to a quantile of the asymptotic distribution of the M-S or K-S statistic. The discrepancy method considered earlier for light-tailed pdfs is investigated now for heavy-tailed pdfs.
KeywordsHeavy-tailed density Kernel estimator Bandwidth Discrepancy method
This work was supported in part by the Russian Foundation for Basic Research, grant 13-08-00744.
- 2.Bolshev, L.N., Smirnov, N.V.: Tables of Mathematical Statistics. Nauka, Moscow (1965) (in Russian)Google Scholar
- 4.Bowman, A.W., Azzalini, A.: Applied Smoothing Techniques for Data Analysis. Oxford University Press (1997)Google Scholar
- 7.Devroye, L., Györfi, L.: Nonparametric Density Estimation. The \(L_1\) View. Wiley, New York (1985)Google Scholar
- 8.Dobrovidov A.V., Markovich L.A.: Data-driven bandwidth choice for gamma kernel estimates of density derivatives on the positive semi-axis. In: Proceedings of IFAC International Workshop on Adaptation and Learning in Control and Signal Processing, pp. 500–505. Caen, France, 3–5 July (2013). doi: 10.3182/20130703-3-FR-4038.00086, arXiv:1401.6801 Google Scholar
- 11.Markovich, L.A.: Nonparametric gamma kernel estimators of density derivatives on positive semiaxis by dependent data. RevStat Stat. J. (2016). arXiv:1401.6783 (in appear)
- 12.Markovich, L.A.: Nonparametric estimation of multivariate density and its derivative by dependent data using gamma kernels. Submitted J. Nonparametric Stat. (2016). arXiv:1410.2507
- 13.Markovich, N.M.: Experimental analysis of nonparametric probability density estimates and of methods for smoothing them. Autom. Remote Control 50, 941–948 (1989)Google Scholar
- 15.Markovich, N.M.: Nonparametric estimation of extremal index using discrepancy method. In: Proceedings of the X International Conference System Identification and Control Problems SICPRO-2015, pp. 160–168. V.A. Trapeznikov Institute of Control Sciences, Moscow, 26–29 Jan 2015. ISBN: 978-5-91450-162-1Google Scholar
- 16.Martynov, G.V.: Omega-Square Criteria. Nauka, Moscow (1978) (in Russian)Google Scholar
- 19.Scott, D.W.: Multivariate Density Estimation. Theory, Practice, and Visualization. Wiley, New York, Chichester (1992)Google Scholar
- 22.Zambom, A.Z., Dias, R.: A Review of Kernel Density Estimation with Applications to Econometrics (2012). arXiv:1212.2812