Computer investigation of the Rate of Convergence of Lepage Type Series to
-Stable Random Variables
Authors:
Aleksander Janicki a;
Piotr Kokoszka a
| Affiliation: | a Institute of Mathematics, Technical University, Wroc aw, Poland |
DOI:
10.1080/02331889208802383
Publication Frequency:
6 issues per year
Subjects:
Mathematical Statistics;
Statistical Theory & Methods;
Statistics;
Statistics for the Biological Sciences;
Stochastic Models & Processes;
Formats available:
PDF
(English)
View Article:
View Article (PDF)
Abstract
The aim of the paper is to present the results of computer experiments that allowed us to investigate some of the properties of LePage series representations of
-stable random variables and to check their usefulness to computer simulation of L vy motion. We base our study on the explicitly obtained formula for the expectation , where , with τj's denoting the arrival times of a Poisson process and appropriately chosen discrete random variables, and where X denotes an a.s. limit of the sequence Xn . We conclude that as far as computer simulations are concerned the sums Xn are not the most suitable means of approximation of X.
|
| Keywords: Stable random variables; LePage series representations; computer simulations of random variables and processes |
| AMS 1980 subject classification: Primary 62E10; secondary 65U05 |
| view references (10) |

Download Citation


aw, Poland
-stable random variables and to check their usefulness to computer simulation of L
vy motion. We base our study on the explicitly obtained formula for the expectation 


Xn
. We conclude that as far as computer simulations are concerned the sums Xn are not the most suitable means of approximation of X.
CiteULike
Del.icio.us
BibSonomy
Connotea