On the stable-number generator design with characteristic exponent greater than one | Vestnik Tomskogo gosudarstvennogo universiteta. Upravlenie, vychislitelnaja tehnika i informatika – Tomsk State University Journal of Control and Computer Science. 2013. № 4(25).

On the stable-number generator design with characteristic exponent greater than one

The stable distribution random number generator with characteristic factor greater than one is developed. Simulation algorithm is based on the General Central Limit Theorem according to that stable distributions be limit distributions for normalised and centralised sum of independent and identically distributed random variables from the domain of attraction for stable distributions: S = +... + ——»Y if n a e (1,2) . bn As summands we use the Pareto mixtures whose carriers belong to the positive and negative semi-axes. The limit equation for characteristic function of sum S was obtained. The relation for parameters of stable distribution in (A) parameterization and Pareto distribution parameters was obtained, as well as the expression for normalising factor b. Resulted characteristics - values for three error functionals corresponding to Pareto mixtures with specified mixture parameters (r, l) and (a, p ) parameters are given in the table. The simulation results are illustrated by the figures 1 and 2. Note that limit equations for Sn, , characteristic function and corresponding expression for b when a^2 do not provide sufficient approximation accuracy.

Download file

Counter downloads: 348

Keywords

stable distributions, Pareto distribution, modeling random variable, распределение Парето, распределения, устойчивые, моделирование случайной величины

Authors

Name	Organization	E-mail
Bagrova I.A.	Tver State University	inna@tversu.ru

Всего: 1

References

Nolan J.P. Numerical calculation of stable densities and distribution functions // Commun. Statist. Stochastic Models. 1997. V.13. P. 759-774.

Chambers J., Mallows C., StuckB. A method for simulating stable random variables // Journal of the American Statistical Association. Theory and Methods Section. 1976. V. 71. No. 354. P. 340-344.

Uchaikin V.V., Zolotarev V.M. Chance and Stability. Stable Distributions and their Applications. Utrecht: VSP, 1999. 594 p.

Janicki A., Weron A. Simulation and Chaotic Behavior of -Stable Stochastic Processes. New York: Marcel Dekker, 1994. 355 p.

Денисов В.И., Тимофеев В.С. Устойчивые распределения и оценивание параметров регрессионных зависимостей // Известия Томского политехнического университета. Томск: Изд-во ТПУ, 2011. Т. 318. № 2. С. 10-15.

Mittnik S., Rachev S., and Schwartz E. Value-at-risk and asset allocation with stable return distributions // Allgemeines Statistisches Archiv. 2002. V. 86. No. 1. P. 53-68.

Маслов О.Н. Устойчивые распределения и их применение в радиотехнике. М: Радио и связь, 1994. 152 с.

Samorodnitsky G. and Taqqu M.S. Stable Non-Gaussian Random Processes. New York: Chapman and Hall, 1994. 632 p.

Барду Ф., Бушо Ж.-Ф., Аспе А., Коэн-Таннуджи К. Статистика Леви и лазерное охлаждение. Как редкие события останавливают атомы: пер. с англ. / под ред. В.П. Яковлева. М.: ФИЗМАТЛИТ, 2006. 216 с.

Rachev S., Mittnik S. Stable Paretian Models in Finance. Wiley, 2000. 855 p.

Золотарев В.М. Одномерные устойчивые распределения. М.: Наука, 1983. 304 с.