Previous |  Up |  Next

Article

Title: Nonparametric estimations of non-negative random variables distributions (English)
Author: Vávra, František
Author: Nový, Pavel
Author: Mašková, Hana
Author: Kotlíková, Michala
Author: Zmrhal, David
Language: English
Journal: Kybernetika
ISSN: 0023-5954
Volume: 39
Issue: 3
Year: 2003
Pages: [341]-346
Summary lang: English
.
Category: math
.
Summary: The problem of estimation of distribution functions or fractiles of non- negative random variables often occurs in the tasks of risk evaluation. There are many parametric models, however sometimes we need to know also some information about the shape and the type of the distribution. Unfortunately, classical approaches based on kernel approximations with a symmetric kernel do not give any guarantee of non-negativity for the low number of observations. In this note a heuristic approach, based on the assumption that non-negative distributions can be also approximated by means of kernels which are defined only on the positive real numbers, is discussed. (English)
Keyword: distribution function
Keyword: kernelapproximation
Keyword: non-negative random variable
MSC: 62G05
MSC: 62G07
idZBL: Zbl 1245.62033
idMR: MR1995738
.
Date available: 2009-09-24T19:54:32Z
Last updated: 2015-03-23
Stable URL: http://hdl.handle.net/10338.dmlcz/135536
.
Reference: [1] German H.: Learning about risk: some lessons from insurance.European Finance Review 2 (1999), 113–124 10.1023/A:1009835429630
Reference: [2] Devroye L., Györfi L.: Nonparametric Density Estimation the $L_1$-view.Wiley, New York 1985 MR 0780746
Reference: [3] Albrecher H.: Dependent Risks and Ruin Probabilities in Insurance.Interim Report, IIASA, IR 98 072
Reference: [4] Rao C. R.: Linear Statistical Inference and its Applications (Czech translation).Academia, Praha 1978
Reference: [5] Rényi A.: Theory of Probability (Czech translation).Academia, Praha 1972 MR 0350789
.

Files

Files Size Format View
Kybernetika_39-2003-3_10.pdf 865.2Kb application/pdf View/Open
Back to standard record
Partner of
EuDML logo