Abstract
We address the estimation of extreme level curves of heavy-tailed distributions. This problem is equivalent to estimating quantiles when covariate information is available and when their order converges to one as the sample size increases. We show that, under some conditions, these so-called “extreme conditional quantiles” can still be estimated through a kernel estimator of the conditional survival function. Sufficient conditions on the rate of convergence of their order to one are provided to obtain asymptotically Gaussian distributed estimators. Making use of this result, some kernel estimators of the conditional tail-index are introduced and a Weissman type estimator is derived, permitting to estimate extreme conditional quantiles of arbitrary large order. These results are illustrated through simulated and real datasets.
Similar content being viewed by others
References
Alves MIF, Gomes MI, de Haan L (2003) A new class of semi-parametric estimators of the second order parameter. Port Math 60:193–214
Alves MIF, de Haan L, Lin T (2003) Estimation of the parameter controlling the speed of convergence in extreme value theory. Math Methods Stat 12:155–176
Beirlant J, Goegebeur Y (2003) Regression with response distributions of Pareto-type. Comput Stat Data Anal, 42:595–619
Berlinet A, Gannoun A, Matzner-Løber E (2001) Asymptotic normality of convergent estimates of conditional quantiles. Statistics 35:139–169
Bernard-Michel C, Douté S, Fauvel M, Gardes L, Girard S (2009) Retrieval of Mars surface physical properties from OMEGA hyperspectral images using Regularized Sliced Inverse Regression. J Geophys Res Planets 114:E06005
Bernard-Michel C, Gardes L, Girard S (2009) Gaussian regularized sliced inverse regression. Stat Comput 19:85–98
Bingham NH, Goldie CM, Teugels JL (1987) Regular variation. Cambridge University Press, Cambridge
Chavez-Demoulin V, Davison AC (2005) Generalized additive modelling of sample extremes. J R Stat Soc Ser C 54:207–222
Collomb G (1976) Estimation non paramétrique de la régression par la méthode du noyau. PhD thesis, Université Paul Sabatier de Toulouse
Davison AC, Ramesh NI (2000) Local likelihood smoothing of sample extremes. J R Stat Soc, Ser B 62:191–208
Davison AC, Smith RL (1990) Models for exceedances over high thresholds. J R Stat Soc, Ser B 52:393–442
Dekkers A, de Haan L (1989) On the estimation of the extreme-value index and large quantile estimation. Ann Stat 17:1795–1832
Einmahl JHJ (1990) The empirical distribution function as a tail estimator. Stat Neerl 44:79–82
Embrechts P, Klüppelberg C, Mikosch T (1997) Modelling extremal events. Springer, Berlin
Ferraty F, Vieu P (2005) Nonparametric modelling for functional data. Springer, Berlin
Gannoun, A (1990) Estimation non paramétrique de la médiane conditionnelle, médianogramme et méthode du noyau. Publ Inst Stat Univ Paris XXXXVI:11–22
Gannoun A, Girard S, Guinot C, Saracco J (2002) Reference ranges based on nonparametric quantile regression. Stat Med 21:3119–3135
Gardes L (2002) Estimating the support of a Poisson process via the Faber-Shauder basis and extreme values. Publ Inst Stat Univ Paris XXXXVI:43–72
Gardes L, Girard S (2008) A moving window approach for nonparametric estimation of the conditional tail index. J Multivar Anal 99:2368–2388
Gardes L, Girard S, Lekina A (2010) Functional nonparametric estimation of conditional extreme quantiles. J Multivar Anal 101:419–433
Geffroy J (1964) Sur un problème d’estimation géométrique. Publ Inst Stat Univ Paris XIII:191–210
Gijbels I, Peng L (2000) Estimation of a support curve via order statistics. Extremes 3:251–277
Girard S, Jacob P (2004) Extreme values and kernel estimates of point processes boundaries. ESAIM: Probab Stat 8:150–168
Girard S, Jacob P (2008) Frontier estimation via kernel regression on high power-transformed data. J Multivar Anal 99:403–420
Girard S, Menneteau L (2005) Central limit theorems for smoothed extreme value estimates of point processes boundaries. J Stat Plan Inference 135(2):433–460
Gomes MI, Martins MJ, Neves M (2000) Semi-parametric estimation of the second order parameter, asymptotic and finite sample behaviour. Extremes 3:207–229
de Haan L, Ferreira A (2006) Extreme value theory: an introduction. Springer series in operations research and financial engineering. Springer, Berlin
Hall P, Tajvidi N (2000) Nonparametric analysis of temporal trend when fitting parametric models to extreme-value data. Stat Sci 15:153–167
Hall P, Nussbaum M, Stern S (1997) On the estimation of a support curve of indeterminate sharpness. J Multivar Anal 62(2):204–232
Härdle W, Park BU, Tsybakov AB (1995) Estimation of a non sharp support boundaries. J Multivar Anal 43:205–218
Hill BM (1975) A simple general approach to inference about the tail of a distribution. Ann Stat 3:1163–1174
Korostelev AP, Tsybakov AB (1993) Minimax theory of image reconstruction. Lecture notes in statistics, vol 82. Springer, New York
Meligkotsidou L, Vrontos I, Vrontos S (2009) Quantile regression analysis of hedge fund strategies. J Empir Finance 16:264–279
Menneteau L (2008) Multidimensional limit theorems for smoothed extreme value estimates of point processes boundaries. ESAIM: Probab Stat 12:273–307
Pickands J (1975) Statistical inference using extreme order statistics. Ann Stat 3:119–131
Roussas GG (1969) Nonparametric estimation of the transition distribution function of a Markov process. Ann Math Stat 40:1386–1400
Samanta T (1989) Non-parametric estimation of conditional quantiles. Stat Probab Lett 7:407–412
Smith RL (1989) Extreme value analysis of environmental time series: an application to trend detection in ground-level ozone (with discussion). Stat Sci 4:367–393
Stone CJ (1977) Consistent nonparametric regression (with discussion). Ann Stat 5:595–645
Stute W (1986) Conditional empirical processes. Ann Stat 14:638–647
Weissman I (1978) Estimation of parameters and large quantiles based on the k largest observations. J Am Stat Assoc 73:812–815
Yao Q (1999) Conditional predictive regions for stochastic processes. Technical report, University of Kent at Canterbury
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Daouia, A., Gardes, L., Girard, S. et al. Kernel estimators of extreme level curves. TEST 20, 311–333 (2011). https://doi.org/10.1007/s11749-010-0196-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11749-010-0196-0