A Unified Approach to the Moments Based Distribution Estimation – Unbounded Support

  • Árpád Tari
  • Miklós Telek
  • Peter Buchholz
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3670)

Abstract

The problem of moments has been studied for more than a century. This paper discusses a practical issue related to the problem of moments namely the bounding of a distribution based on a given number of moments. The presented approach is unified in the sense that all measures of interests are provided as a quadratic expression of the same Hankel-matrix.

Application examples indicate the importance of the presented approach.

Keywords

reduced moment problem moments based distribution bounding Hankel matrix 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Akhiezer, N.I.: The classical moment problem and some related questions in analysis. Hafner publishing company, New York (1965) (translation of Н. И. Ахиезер Класическая Проблема Моментов и Некоторые Волросы Анализа published by Государственное Издательство физико Математической Литературы, Moscow, 1961) Google Scholar
  2. 2.
    Donatiello, L., Grassi, V.: On evaluating the cumulative performance distribution of fault-tolerant computer systems. IEEE Transactions on Computers (1991)Google Scholar
  3. 3.
    Fodor, G., Rácz, S., Telek, M.: On providing blocking probability- and throughput guarantees in a multi-service environment. International Journal of Communication Systems 15(4), 257–285 (2002)MATHCrossRefGoogle Scholar
  4. 4.
    Horváth, A., Telek, M., Horváth, G.: Analysis of inhomogeneous Markov reward models. Linear algebra and its application 386, 383–405 (2004)MATHCrossRefGoogle Scholar
  5. 5.
    Horváth, G., Rácz, S., Tari, Á., Telek, M.: Evaluation of reward analysis methods with MRMSolve 2.0. In: 1st International Conference on Quantitative Evaluation of Systems (QEST) 2004, Twente, The Netherlands, September 2004, pp. 165–174. IEEE CS Press, Los Alamitos (2004)CrossRefGoogle Scholar
  6. 6.
    van de Liefvoort, A.: The moment problem for continuous distributions. Technical report, University of Missouri, WP-CM-1990-02, Kansas City (1990)Google Scholar
  7. 7.
    Nabli, H., Sericola, B.: Performability analysis: a new algorithm. IEEE Transactions on Computers 45, 491–494 (1996)MATHCrossRefGoogle Scholar
  8. 8.
    Press, W.H., Flannery, B.P., Teukolsky, S.A., Vetterling, W.T.: Numerical Recipes in C: The Art of Scientific Computing. Cambridge University Press, Cambridge (1993), http://lib-www.lanl.gov/numerical/bookcpdf.html Google Scholar
  9. 9.
    Rácz, S.: Numerical analysis of communication systems through Markov reward models. PhD thesis, Technical University of Budapest (2000)Google Scholar
  10. 10.
    Rácz, S., Telek, M.: Performability analysis of Markov reward models with rate and impulse reward. In: Silva, M., Plateau, B., Stewart, W. (eds.) Int. Conf. on Numerical solution of Markov chains, Zaragoza, Spain, pp. 169–187 (1999)Google Scholar
  11. 11.
    Rózsa, P.: Lineáris algebra és alkalmazásai. Tankönyvkiadó (1991) (in Hungarian)Google Scholar
  12. 12.
    Shohat, J.A., Tamarkin, D.J.: The problem of moments. Americal Mathematical Society, Providence, Rhode Island, Mathematical surveys (1946)Google Scholar
  13. 13.
    de Souza e Silva, E., Gail, H.R.: Calculating cumulative operational time distributions of repairable computer systems. IEEE Transactions on Computers C-35, 322–332 (1986)CrossRefGoogle Scholar
  14. 14.
    Stieltjes, T.: Reserches sur les fractions continues. Ann. Fac. Sci. Univ. Toulouse 2, 1–122 (1894) (in French)Google Scholar
  15. 15.
    Szegö, G.: Orthogonal polynomials. American Mathematical Society, Providence (1939)Google Scholar
  16. 16.
    Tagliani, A.: Existence and stability of a discrete probability distribution by maximum entropy approach. Applied Mathematics and Computation 110, 105–114 (2000)MATHCrossRefMathSciNetGoogle Scholar
  17. 17.
    Telek, M., Rácz, S.: Numerical analysis of large Markovian reward models. Performance Evaluation 36&37, 95–114 (1999)CrossRefGoogle Scholar
  18. 18.
    Weisstein, E.W.: Positive definite matrix, http://mathworld.wolfram.com/PositiveDefiniteMatrix.html

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Árpád Tari
    • 1
  • Miklós Telek
    • 2
  • Peter Buchholz
    • 1
  1. 1.Universität DortmundGermany
  2. 2.Budapest University of Technology and EconomicsHungary

Personalised recommendations