Abstract
This paper is concerned with extreme value density estimation. The generalized Pareto distribution (GPD) beyond a given threshold is combined with a nonparametric estimation approach below the threshold. This semiparametric setup is shown to generalize a few existing approaches and enables density estimation over the complete sample space. Estimation is performed via the Bayesian paradigm, which helps identify model components. Estimation of all model parameters, including the threshold and higher quantiles, and prediction for future observations is provided. Simulation studies suggest a few useful guidelines to evaluate the relevance of the proposed procedures. They also provide empirical evidence about the improvement of the proposed methodology over existing approaches. Models are then applied to environmental data sets. The paper is concluded with a few directions for future work.
Similar content being viewed by others
References
Asmussen, S.: Applied Probability and Queues. Wiley, New York (1987)
Behrens, C., Gamerman, D., Lopes, H.F.: Bayesian analysis of extreme events with threshold estimation. Stat. Model. 4, 227–244 (2004)
Bermudez, P., Turkman, M.A., Turkman, K.F.: A predictive approach to tail probability estimation. Extremes 4, 295–314 (2001)
Cabras, S., Castellanos, M.A., Gamerman, D.: A default Bayesian approach for regression on extremes. Stat. Model. (2011, accepted)
Castellanos, M.A., Cabras, S.: A default Bayesian procedure for the generalized Pareto distribution. J. Stat. Plan. Inference 137, 473–483 (2007)
Coles, S.G.: Extreme Value Theory an Applications. Kluver Academic, Dordrecht (2001)
Coles, S.G., Tawn, J.A.: A Bayesian analysis of extreme rainfall data. Appl. Stat. 45, 463–478 (1996)
Cunnane, C.: Note on the Poisson assumption in partial duration series model. Water Resour. Res. 15, 489–494 (1979)
Dalal, S., Hall, W.: Approximating priors by mixtures of natural conjugate priors. J. R. Stat. Soc., Ser. B 45, 278–286 (1983)
Davison, A.C., Smith, R.L.: Models for exceedances over high thresholds (with discussion). J. R. Stat. Soc., Ser. B 52, 393–342 (1990)
Dey, D., Kuo, L., Sahu, S.: A Bayesian predictive approach to determining the number of components in a mixture distribution. Stat. Comput. 5, 297–305 (1995)
Diebolt, J., Robert, C.: Estimation of finite mixture distributions by Bayesian sampling. J. R. Stat. Soc., Ser. B 56, 363–375 (1994)
Diebolt, J., El-Aroui, M., Garrido, M., Girard, S.: Quasi-conjugate Bayes estimates for gpd parameters and application to heavy tails modelling. Extremes 1, 57–78 (2005)
Doornik, JA: Ox: Object Oriented Matrix Programming, 4.1 console version. Nuffield College, Oxford University, London (1996)
Embrechts, P., Küppelberg, C., Mikosch, T.: Modelling Extremal Events for Insurance and Finance. Springer, New York (1997)
Fisher, R.A., Tippett, L.H.C.: On the estimation of the frequency distributions of the largest and smallest number of a sample. Proc. Camb. Philos. Soc. 24, 180–190 (1928)
Frigessi, A., Haug, O., Rue, H.: A dynamic mixture model for unsupervised tail estimation without threshold selection. Extremes 5, 219–235 (2002)
Frühwirth-Schnatter, S.: Markov chain Monte Carlo estimation of classical and dynamic switching and mixture models. J. Am. Stat. Assoc. 96, 194–209 (2001)
Gamerman, D., Lopes, H.F.: Markov Chain Monte Carlo: Stochastic Simulation for Bayesian Inference, 2nd ed. Chapman and Hall/CRC, Baton Rouge (2006)
Gramacy, R., Lee, K.: Bayesian treed Gaussian process models with an application to computer modeling. J. Am. Stat. Assoc. 103, 1119–1130 (2008)
Jenkinson, A.F.: The frequency distribution of the annual maximum (or minimum) values of meteorological events. Q. J. R. Meteorol. Soc. 81, 158–171 (1955)
Lopes, H.F., Nascimento, F.F., Gamerman, D.: Generalized Pareto models with time-varying tail behavior. Technical Report LES:UFRJ, in preparation (2011)
von Mises, R.: La distribution de la plus grande de nvaleurs. Am. Math. Soc. 2, 271–294 (1954)
Nascimento, F.F., Gamerman, D., Lopes, H.F.: Regression models for exceedance data via the full likelihood. Environ. Ecol. Stat. (2011, to appear)
Pickands, J.: Statistical inference using extreme order statistics. Ann. Stat. 3, 119–131 (1975)
Richardson, S., Green, P.: On Bayesian analysis of mixtures with an unknown number of components. J. R. Stat. Soc., Ser. B 59, 731–792 (1997)
Roberts, G.O., Rosenthal, J.S.: Examples of adaptive mcmc. Journal of Computation and Graphical. Statistics 18, 349–367 (2009)
Roeder, K., Wasserman, L.: Practical Bayesian density estimation using mixtures of normals. J. Am. Stat. Assoc. 92, 894–902 (1997)
Schwarz, G.: Estimating the dimension of a model. Ann. Stat. 6, 461–464 (1978)
Smith, R.L.: Threshold models for sample extremes. Statistical extremes and applications 621–638 (1984)
Spiegelhalter, D.J., Best, N.G., Carlin, B.P., Linde, A.: Bayesian measures of model complexity and fit. J. R. Stat. Soc. B 64, 583–639 (2002)
Tancredi, A., Anderson, C., O’Hagan, A.: Accounting for threshold uncertainty in extreme value estimation. Extremes 9, 87–106 (2006)
Titterington, D., Smith, A.F.M., Makov, U.: Statistical Analysis of Finite Mixture Distributions. Wiley, New York (1985)
Wiper, M., Rios Insua, D., Ruggeri, F.: Mixtures of gamma distributions with applications. J. Comput. Graph. Stat. 10, 440–454 (2001)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
do Nascimento, F.F., Gamerman, D. & Lopes, H.F. A semiparametric Bayesian approach to extreme value estimation. Stat Comput 22, 661–675 (2012). https://doi.org/10.1007/s11222-011-9270-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-011-9270-z