Kybernetika 48 no. 1, 31-49, 2012

On the extremal behavior of a Pareto process: an alternative for ARMAX modeling

Marta Ferreira


In what concerns extreme values modeling, heavy tailed autoregressive processes defined with the minimum or maximum operator have proved to be good alternatives to classical linear ARMA with heavy tailed marginals (Davis and Resnick [8], Ferreira and Canto e Castro [13]). In this paper we present a complete characterization of the tail behavior of the autoregressive Pareto process known as Yeh-Arnold-Robertson Pareto(III) (Yeh et al. [32]). We shall see that it is quite similar to the first order max-autoregressive ARMAX, but has a more robust parameter estimation procedure, being therefore more attractive for modeling purposes. Consistency and asymptotic normality of the presented estimators will also be stated.


Markov chains, tail dependence, extreme value theory, autoregressive processes


60G70, 60J20


  1. M. T. Alpuim: An extremal markovian sequence. J. Appl. Probab. 26 (1989), 219-232.   CrossRef
  2. B. C. Arnold: Pareto Distributions. International Cooperative Publishing House, Fairland 1983.   CrossRef
  3. B. C. Arnold: Pareto processes. In: Handbook of Statistics (D. N. Shanbhag and C. R. Rao, eds.), Elsevier Science B.V. 2001, Vol. 19.   CrossRef
  4. S. Asmussen: Applied Probability and Queues. John Wiley \et Sons, Chichester 1987.   CrossRef
  5. L. Canto e Castro: Sobre a Teoria Assintótica de Extremos. Ph.D. Thesis, FCUL 1992.   CrossRef
  6. M. R. Chernick: A limit theorem for the maximum of autoregressive processes with uniform marginal distribution. Ann. Probab. 9 (1981), 145-149.   CrossRef
  7. M. R. Chernick, T. Hsing and W. P. McCormick: Calculating the extremal index for a class of stationary sequences. Adv. Probab. 23 (1991), 835-850.   CrossRef
  8. R. Davis and S. Resnick: Basic properties and prediction of max-ARMA processes. Adv. Appl. Probab. 21 (1989), 781-803.   CrossRef
  9. D. J. Daley and J. Haslett: A thermal energy storage process with controlled input. Adv. Appl. Probab. 14 (1982), 257-271.   CrossRef
  10. A. L. M. Dekkers, J. H. J. Einmahl and L. de Haan: A moment estimator for the index of an extreme value distribution. Ann. Statist. 17 (1989), 1833-1855.   CrossRef
  11. H. Drees: Extreme quantile estimation for dependent data with applications to finance. Bernoulli 9 (2003), 617-657.   CrossRef
  12. H. Ferreira: The upcrossings index and the extremal index. J. Appl. Probab. 43(4) (2006), 927-937.   CrossRef
  13. M. Ferreira and L. Canto e Castro: Modeling rare events through a $p$RARMAX process. J. Statist. Plann. Inference 140 (2010), 11, 3552-3566.   CrossRef
  14. M. Ferreira and H. Ferreira: On extremal dependence: some contributions. (In press).   CrossRef
  15. I. S. Helland and T. S. Nilsen: On a general random exchange model. J. Appl. Probab. 13 (1976), 781-790.   CrossRef
  16. B. M. Hill: A simple general approach to inference about the tail of a distribution. Ann. Statist. 3 (1975), 1163-1174.   CrossRef
  17. J. R. M. Hosking and J. R. Wallis: Parameter and quantile estimation for the generalized Pareto distribution. Technometrics 29 (1987), 339-349.   CrossRef
  18. T. Hsing, J. Hüsler and M. R. Leadbetter: On the exceedance point process for a stationary sequence. Probab. Theory Related Fields 78 (1988), 97-112.   CrossRef
  19. J. Klotz: Statistical inference in Bernoulli trials with dependence. Ann. Statist. 1 (1973), 373-379.   CrossRef
  20. M. R. Leadbetter: On extreme values in stationary sequences. Z. Wahrsch. verw. Gebiete 28 (1974), 289-303.   CrossRef
  21. M. R. Leadbetter, G. Lindgren and H. Rootzén: Extremes and Related Properties of Random Sequences and Processes. Springer-Verlag, New York 1983.   CrossRef
  22. M. R. Leadbetter and S. Nandagopalan: On exceedance point processes for stationary sequences under mild oscillation restrictions. In: Extreme Value Theory (J. Hüsler and R.-D. Reiss, eds.), Springer-Verlag 1989, pp. 69-80.   CrossRef
  23. A. V. Lebedev: Statistical analysis of first-order MARMA processes. Mat. Zametki 83 (2008), 4, 552-558.   CrossRef
  24. A. Ledford and J. A. Tawn: Statistics for near independence in multivariate extreme values. Biometrika 83 (1996), 169-187.   CrossRef
  25. A. Ledford and J. A. Tawn: Modelling dependence within joint tail regions. J. Royal Statist. Soc. Ser. B 59 (1997), 475-499.   CrossRef
  26. V. Pareto: Cours d'economie Politique. F. Rouge, Lausanne Vol. II., 1897.   CrossRef
  27. J. Pickands III: Statistical inference using extreme order statistics. Ann. Statist. 3 (1975), 119-131.   CrossRef
  28. S. Resnick and C. St\v{a}ric\v{a}: Consistency of Hill's estimator for dependent data. J. Appl. Probab. 32 (1995), 139-167.   CrossRef
  29. S. Resnick and C. St\v{a}ric\v{a}: Tail index estimation for dependent data. Ann. Appl. Probab. 8 (1998), 4, 1156-1183.   CrossRef
  30. H. Rootzén, M. R. Leadbetter and L. de Haan: Tail and Quantile Estimation for Strongly Mixing Stationary Sequences. Technical Report, UNC Center for Stochastic Processes, 1990.   CrossRef
  31. R. L. Smith: Estimating tails of probability distributions. Ann. Statist. 15 (1987), 1174-1207.   CrossRef
  32. H. C. Yeh, B. C. Arnold and C. A. Robertson: Pareto processes. J. Appl. Probab. 25 (1988), 291-301.   CrossRef
  33. Z. Zhang and R. L. Smith: Modelling Financial Time Series Data as Moving Maxima Processes. Technical Report Dept. Stat. (Univ. North Carolina, Chapel Hill, NC, 2001); papers/RLS_Papers.html.   CrossRef