Geodesic
Approximated maximum likelihood estimation of parameters of discrete stable family
Kybernetika, Tome 50 (2014) no. 6, pp. 1065-1076
Cet article a éte moissonné depuis la source Czech Digital Mathematics Library

Voir la notice de l'article

In this article we propose a method of parameters estimation for the class of discrete stable laws. Discrete stable distributions form a discrete analogy to classical stable distributions and share many interesting properties with them such as heavy tails and skewness. Similarly as stable laws discrete stable distributions are defined through characteristic function and do not posses a probability mass function in closed form. This inhibits the use of classical estimation methods such as maximum likelihood and other approach has to be applied. We depart from the $\mathcal{H}$-method of maximum likelihood suggested by Kagan (1976) where the likelihood function is replaced by a function called informant which is an approximation of the likelihood function in some Hilbert space. For this method only some functionals of the distribution are required, such as probability generating function or characteristic function. We adopt this method for the case of discrete stable distributions and in a simulation study show the performance of this method.
In this article we propose a method of parameters estimation for the class of discrete stable laws. Discrete stable distributions form a discrete analogy to classical stable distributions and share many interesting properties with them such as heavy tails and skewness. Similarly as stable laws discrete stable distributions are defined through characteristic function and do not posses a probability mass function in closed form. This inhibits the use of classical estimation methods such as maximum likelihood and other approach has to be applied. We depart from the $\mathcal{H}$-method of maximum likelihood suggested by Kagan (1976) where the likelihood function is replaced by a function called informant which is an approximation of the likelihood function in some Hilbert space. For this method only some functionals of the distribution are required, such as probability generating function or characteristic function. We adopt this method for the case of discrete stable distributions and in a simulation study show the performance of this method.
DOI : 10.14736/kyb-2014-6-1065
Classification : 60E07, 60E10, 62F12, 62G05, 65C60
Keywords: discrete stable distribution; parameter estimation; maximum likelihood 
@article{10_14736_kyb_2014_6_1065,
     author = {Sl\'amov\'a, Lenka and Klebanov, Lev B.},
     title = {Approximated maximum likelihood estimation of parameters of discrete stable family},
     journal = {Kybernetika},
     pages = {1065--1076},
     year = {2014},
     volume = {50},
     number = {6},
     doi = {10.14736/kyb-2014-6-1065},
     mrnumber = {3301786},
     zbl = {1308.60022},
     language = {en},
     url = {http://geodesic.mathdoc.fr/articles/10.14736/kyb-2014-6-1065/}
}
TY  - JOUR
AU  - Slámová, Lenka
AU  - Klebanov, Lev B.
TI  - Approximated maximum likelihood estimation of parameters of discrete stable family
JO  - Kybernetika
PY  - 2014
SP  - 1065
EP  - 1076
VL  - 50
IS  - 6
UR  - http://geodesic.mathdoc.fr/articles/10.14736/kyb-2014-6-1065/
DO  - 10.14736/kyb-2014-6-1065
LA  - en
ID  - 10_14736_kyb_2014_6_1065
ER  - 
%0 Journal Article
%A Slámová, Lenka
%A Klebanov, Lev B.
%T Approximated maximum likelihood estimation of parameters of discrete stable family
%J Kybernetika
%D 2014
%P 1065-1076
%V 50
%N 6
%U http://geodesic.mathdoc.fr/articles/10.14736/kyb-2014-6-1065/
%R 10.14736/kyb-2014-6-1065
%G en
%F 10_14736_kyb_2014_6_1065
Slámová, Lenka; Klebanov, Lev B. Approximated maximum likelihood estimation of parameters of discrete stable family. Kybernetika, Tome 50 (2014) no. 6, pp. 1065-1076. doi: 10.14736/kyb-2014-6-1065

[1] Devroye, L.: A triptych of discrete distributions related to the stable law. Stat. Probab. Lett. 18 (1993), 349-351. | DOI | MR | Zbl

[2] Feuerverger, A., McDunnough, P.: On the efficiency of empirical characteristic function procedure. J. Roy. Stat. Soc. Ser. B 43 (1981), 20-27. | MR

[3] Gerlein, O. V., Kagan, A. M.: Hilbert space methods in classical problems of mathematical statistics. J. Soviet Math. 12 (1979), 184-213. | DOI | Zbl

[4] Kagan, A. M.: Fisher information contained in a finite-dimensional linear space, and a correctly posed version of the method of moments (in Russian). Problemy Peredachi Informatsii 12 (1976), 20-42. | MR

[5] Klebanov, L. B., Melamed, I. A.: Several notes on Fisher information in presence of nuisance parameters. Statistics: J. Theoret. Appl. Stat. 9 (1978), 85-90. | MR | Zbl

[6] Klebanov, L. B., Slámová, L.: Integer valued stable random variables. Stat. Probab. Lett. 83 (2013), 1513-1519. | DOI | MR | Zbl

[7] Slámová, L., Klebanov, L. B.: Modelling financial returns with discrete stable distributions. In: Proc. 30th International Conference Mathematical Methods in Economics (J. Ramík and D. Stavárek, eds.), Silesian University in Opava, School of Business Administration in Karviná, 2012, pp. 805-810.

[8] Steutel, F. W., Harn, K. van: Discrete analogues of self-decomposability and stability. Ann. Probab. 7 (1979), 893-899. | DOI | MR

Cité par Sources :