Parameterizations and Parameters Relations of Stable Distribution

Abstract:

For the problem of various non-Gaussian signals and noises have distinct spiky and impulsive characteristics in practice, which cannot be solved by Gaussian distribution, we focus on the Alpha stable distribution. This paper firstly sets forth the representation of standard parameterization, and gives some properties in this parameterization. Secondly, define a representation of the parameterization for the purpose of being suitable for numerical calculation and modeling, clears the meaning of each parameter, finds relations between parameters in these representations. Finally, analyze and prove the properties equally apply in this parameterization.

• Periodical: Applied Mechanics and Materials (Volumes 543-547)
• Pages: 1721-1727
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.543-547.1721
• Online since: March 2014
• Authors: Chun Li Liu, Hai Long Chen*, Lei Shao, Jun Ting Wang, Cheng Ji You
• Keywords: Characteristic Function, Parameterization, Probability Density Function, Stable Distribution

References

[1] V.M. Zolotarev, One-dimensional Stable Distributions, Amer, Math. Soc. Transl. of Math, Monographs, vol. 65. Amer. Math. Soc., Providence, RI (1986).

[2] J.P. Nolan: Stochastic Models 13 (1997) , p.759.

[3] J.P. Nolan, Maximum likelihood estimation of stable parameters (1997) . Submitted (available on author's webpage).

[4] P. Hall: Bull. London Math. Soc. 13 (1981), p.23.

[5] D. Holt, E. Crow: J. Res. Nat. Bureau Standards, B. Math. Sci. 77B (1973), p.143.

[6] A. Janicki, A. Weron: Stat. Sci. Vol. 9 (1994), p.109.

[7] V.V. Uchaikin, V.M. Zolotarev: Chance and Stability. Utrecht: VSP, Netherlands (1999).