## Bernoulli

• Bernoulli
• Volume 24, Number 2 (2018), 1427-1462.

### Maximum likelihood estimation for the Fréchet distribution based on block maxima extracted from a time series

#### Abstract

The block maxima method in extreme-value analysis proceeds by fitting an extreme-value distribution to a sample of block maxima extracted from an observed stretch of a time series. The method is usually validated under two simplifying assumptions: the block maxima should be distributed exactly according to an extreme-value distribution and the sample of block maxima should be independent. Both assumptions are only approximately true. The present paper validates that the simplifying assumptions can in fact be safely made.

For general triangular arrays of block maxima attracted to the Fréchet distribution, consistency and asymptotic normality is established for the maximum likelihood estimator of the parameters of the limiting Fréchet distribution. The results are specialized to the common setting of block maxima extracted from a strictly stationary time series. The case where the underlying random variables are independent and identically distributed is further worked out in detail. The results are illustrated by theoretical examples and Monte Carlo simulations.

#### Article information

Source
Bernoulli, Volume 24, Number 2 (2018), 1427-1462.

Dates
Revised: September 2016
First available in Project Euclid: 21 September 2017

Permanent link to this document
https://projecteuclid.org/euclid.bj/1505980900

Digital Object Identifier
doi:10.3150/16-BEJ903

Mathematical Reviews number (MathSciNet)
MR3706798

Zentralblatt MATH identifier
06778369

#### Citation

Bücher, Axel; Segers, Johan. Maximum likelihood estimation for the Fréchet distribution based on block maxima extracted from a time series. Bernoulli 24 (2018), no. 2, 1427--1462. doi:10.3150/16-BEJ903. https://projecteuclid.org/euclid.bj/1505980900

#### References

• [1] Balakrishnan, N. and Kateri, M. (2008). On the maximum likelihood estimation of parameters of Weibull distribution based on complete and censored data. Statist. Probab. Lett. 78 2971–2975.
• [2] Bingham, N.H., Goldie, C.M. and Teugels, J.L. (1987). Regular Variation. Encyclopedia of Mathematics and Its Applications 27. Cambridge: Cambridge Univ. Press.
• [3] Bücher, A. and Segers, J. (2014). Extreme value copula estimation based on block maxima of a multivariate stationary time series. Extremes 17 495–528.
• [4] Bücher, A. and Segers, J. (2016). On the maximum likelihood estimator for the Generalized Extreme-Value distribution. Available at arXiv:1601.05702.
• [5] Bücher, A. and Segers, J. (2016). Supplement to “Maximum likelihood estimation for the Fréchet distribution based on block maxima extracted from a time series”. DOI:10.3150/16-BEJ903SUPP.
• [6] Cai, J.-J., de Haan, L. and Zhou, C. (2013). Bias correction in extreme value statistics with index around zero. Extremes 16 173–201.
• [7] de Haan, L. and Ferreira, A. (2006). Extreme Value Theory: An Introduction. Springer Series in Operations Research and Financial Engineering. New York: Springer.
• [8] Dehling, H. and Philipp, W. (2002). Empirical process techniques for dependent data. In Empirical Process Techniques for Dependent Data 3–113. Boston, MA: Birkhäuser Boston.
• [9] Dombry, C. (2015). Existence and consistency of the maximum likelihood estimators for the extreme value index within the block maxima framework. Bernoulli 21 420–436.
• [10] Drees, H. (1998). On smooth statistical tail functionals. Scand. J. Stat. 25 187–210.
• [11] Drees, H. (2000). Weighted approximations of tail processes for $\beta$-mixing random variables. Ann. Appl. Probab. 10 1274–1301.
• [12] Ferreira, A. and de Haan, L. (2015). On the block maxima method in extreme value theory: PWM estimators. Ann. Statist. 43 276–298.
• [13] Gnedenko, B. (1943). Sur la distribution limite du terme maximum d’une série aléatoire. Ann. of Math. (2) 44 423–453.
• [14] Gumbel, E.J. (1958). Statistics of Extremes. New York: Columbia Univ. Press.
• [15] Hosking, J.R.M., Wallis, J.R. and Wood, E.F. (1985). Estimation of the generalized extreme-value distribution by the method of probability-weighted moments. Technometrics 27 251–261.
• [16] Hsing, T. (1991). On tail index estimation using dependent data. Ann. Statist. 19 1547–1569.
• [17] Leadbetter, M.R. (1983). Extremes and local dependence in stationary sequences. Z. Wahrsch. Verw. Gebiete 65 291–306.
• [18] Marohn, F. (1994). On testing the Exponential and Gumbel distribution. In Extreme Value Theory and Applications 159–174. Kluwer Academic.
• [19] Mikosch, T. and Stărică, C. (2000). Limit theory for the sample autocorrelations and extremes of a $\operatorname{GARCH}(1,1)$ process. Ann. Statist. 28 1427–1451.
• [20] Peng, L. (1998). Asymptotically unbiased estimators for the extreme-value index. Statist. Probab. Lett. 38 107–115.
• [21] Pickands, J. III (1975). Statistical inference using extreme order statistics. Ann. Statist. 3 119–131.
• [22] Prescott, P. and Walden, A.T. (1980). Maximum likelihood estimation of the parameters of the generalized extreme-value distribution. Biometrika 67 723–724.
• [23] Resnick, S.I. (1987). Extreme Values, Regular Variation, and Point Processes. Applied Probability. A Series of the Applied Probability Trust 4. New York: Springer.
• [24] Rootzén, H. (2009). Weak convergence of the tail empirical process for dependent sequences. Stochastic Process. Appl. 119 468–490.
• [25] Smith, R.L. (1985). Maximum likelihood estimation in a class of nonregular cases. Biometrika 72 67–90.
• [26] van der Vaart, A.W. (1998). Asymptotic Statistics. Cambridge Series in Statistical and Probabilistic Mathematics 3. Cambridge: Cambridge Univ. Press.

#### Supplemental materials

• Supplement to “Maximum likelihood estimation for the Fréchet distribution based on block maxima extracted from a time series”. The supplementary material Bücher and Segers [5] contains a lemma on moment convergence of block maxima used in the proof of Theorem 4.2 (in Section C), the proof of Lemma 5.1 (in Section D) and the proofs of auxiliary lemmas from Section B (in Section E). Furthermore, we present additional Monte Carlo simulation results to quantify the finite-sample bias and variance of the maximum likelihood estimator (in Section F).