An Algorithm for Fitting Heavy-Tailed Distributions via Generalized Hyperexponentials

References

• Abate J., Whitt W. Computing Laplace transforms for numerical inversion via continued fractions. INFORMS J. Comput. (1999) 11(4):394–405Link, Google Scholar
• Anderson T. W., Darling D. A. Asymptotic theory of certain “goodness of fit” criteria based on stochastic process. Ann. Math. Statist. (1952) 23(2):193–212Crossref, Google Scholar
• Asmussen S., Chakrarthy A. R., Alfa A. S. Phase-type distributions and related point processes: Fitting and recent advances. Matrix-Analytic Methods in Stochastic Models (1996) (Marcel Dekker, New York) 137–149Crossref, Google Scholar
• Asmussen S., Nerman O., Olsson M. Fitting phase-type distributions via the EM algorithm. Scand. J. Statist. (1996) 23(4):419–441Google Scholar
• Bartholomew D. J. Sufficient conditions for a mixture of exponentials to be a probability density function. Ann. Math. Statist. (1969) 40(6):2183–2188Crossref, Google Scholar
• Botta R. F., Harris C. M. Approximation with generalized hyperexponential distributions: Weak convergence results. Queueing Systems (1986) 1(2):169–190Crossref, Google Scholar
• Botta R. F., Marchal W. D., Harris C. M. Characterizations of generalized hyperexponential distribution functions. Stochastic Models (1987) 3(1):115–148Crossref, Google Scholar
• Duffield N. G., Whitt W., Park K., Willinger W. Network design and control using on/off and multilevel source traffic models with heavy-tailed distributions. Self-Similar Network Traffic and Performance Evaluation (2000) (John Wiley & Sons, New York) 421–445Crossref, Google Scholar
• Feldmann A., Whitt W. Fitting mixture of exponentials to long-tail distributions to analyze network performance models. Performance Eval. (1998) 31(3–4):245–279Crossref, Google Scholar
• Harris C. M., Marchal W. G. Distribution estimation using Laplace transforms. INFORMS J. Comput. (1998) 10(4):448–458Link, Google Scholar
• Harris C. M., Marchal W. G., Botta R. F. A note on generalized hyperexponential distributions. Stochastic Models (1992) 8(1):179–191Crossref, Google Scholar
• Hogg R. V., Klugman S. A. On the estimation of long tailed skewed distributions with actuarial applications. J. Econom. (1983) 23(1):91–102Crossref, Google Scholar
• Horvath A., Telek M., Latouche G., Taylor P. Approximating heavy-tailed behavior with phase-type distribution. Advances in Algorithmic Methods for Stochastic Models (2000) (Notable Publications, Neshanic Station, NJ) 191–214Google Scholar
• Johnson N. L., Kotz S., Balakrishnan N.Continuous Univariate Distributions (1994) 12nd ed.(John Wiley & Sons, New York) Google Scholar
• Khayari R. E. A., Sadre R., Haverkort B. R. Fitting World-Wide Web request traces with the EM algorithm. Performance Eval. (2003) 52(2–3):175–191Crossref, Google Scholar
• Nadarajah S., Kotz S. On the Laplace transform of the Pareto distribution. Queueing Systems (2006) 54(4):243–244Crossref, Google Scholar
• Neuts M. F.Matrix-Geometric Solutions in Stochastic Models (1981) (Johns Hopkins University Press, Baltimore) Google Scholar
• Pericchi L. R., Rodriguez-Iturbe I., Atkinson A. C., Fienberg S. E. On the statistical analysis of floods. A Celebration of Statistics: The ISI Centenary Volume (1985) (Springer-Verlag, New York) 511–541Crossref, Google Scholar
• Reiss R.-D., Thomas M.Statistical Analysis of Extreme Values: With Applications to Insurance, Finance, Hydrology and Other Fields (2007) 3rd ed.(Birkhauser Verlag, Basel, Switzerland) Google Scholar
• Shortle J. F. Piecewise-polynomial approximations for heavy-tailed distributions in queueing analysis. Stochastic Models (2005) 21(1):215–234Crossref, Google Scholar
• Shortle J. F., Brill P. H., Fischer M. J., Gross D., Masi D. M. B. An algorithm to compute the waiting time distribution for the M/G/1 queue. INFORMS J. Comput. (2004) 16(2):152–161Link, Google Scholar
• Sigman K. Appendix: A primer on heavy-tailed distributions. Queueing Systems (1999) 33(1–3):261–275Crossref, Google Scholar