Minimum Distance Estimation for the Generalized Pareto Distribution

ABSTRACT The generalized Pareto distribution (GPD) is widely used for extreme values over a threshold. Most existing methods for parameter estimation either perform unsatisfactorily when the shape parameter k is larger than 0.5, or they suffer from heavy computation as the sample size increases. In view of the fact that k > 0.5 is occasionally seen in numerous applications, including two illustrative examples used in this study, we remedy the deficiencies of existing methods by proposing two new estimators for the GPD parameters. The new estimators are inspired by the minimum distance estimation and the M-estimation in the linear regression. Through comprehensive simulation, the estimators are shown to perform well for all values of k under small and moderate sample sizes. They are comparable to the existing methods for k < 0.5 while perform much better for k > 0.5.

[1]  Holger Rootzén,et al.  Extreme Values in Finance, Telecommunications, and the Environment , 2003 .

[2]  W. Rey.  Introduction to Robust and Quasi-Robust Statistical Methods , 1983 .

[3]  M. Kenward,et al.  An Introduction to the Bootstrap , 2007 .

[4]  L. Adkins,et al.  Extreme daily changes in U.S. Dollar London inter-bank offer rates , 2008 .

[5]  J. Hart,et al.  Nonparametric Estimation of Distributions in Random Effects Models , 2011 .

[6]  V. Yohai,et al.  High Breakdown-Point Estimates of Regression by Means of the Minimization of an Efficient Scale , 1988 .

[7]  Jerald F. Lawless,et al.  Statistical Methods in Reliability , 1983 .

[8]  Louis A. Jaeckel.  Estimating Regression Coefficients by Minimizing the Dispersion of the Residuals , 1972 .

[9]  J. Hosking,et al.  Parameter and quantile estimation for the generalized pareto distribution , 1987 .

[10]  Sonja Kuhnt,et al.  Goodness‐of‐Fit Tests , 2014 .

[11]  S. Grimshaw.  Computing Maximum Likelihood Estimates for the Generalized Pareto Distribution , 1993 .

[12]  Alberto Luceño,et al.  Fitting the generalized Pareto distribution to data using maximum goodness-of-fit estimators , 2006, Comput. Stat. Data Anal..

[13]  P. J. Huber.  The 1972 Wald Lecture Robust Statistics: A Review , 1972 .

[14]  W. Nelson.  Statistical Methods for Reliability Data , 1998 .

[15]  Zhi-Sheng Ye,et al.  Estimation of Field Reliability Based on Aggregate Lifetime Data , 2017, Technometrics.

[16]  Richard L. Smith.  Threshold Methods for Sample Extremes , 1984 .

[17]  M. Zagórski,et al.  Analysis of the turbine steady-state data by means of Generalized Pareto Distribution , 2007 .

[18]  Jin Zhang.  Powerful goodness‐of‐fit tests based on the likelihood ratio , 2002 .

[19]  Holger Rootzén,et al.  Tail Estimation for Window-Censored Processes , 2016, Technometrics.

[20]  J. Wolfowitz.  The Minimum Distance Method , 1957 .

[21]  J. Pickands.  Statistical Inference Using Extreme Order Statistics , 1975 .

[22]  D. F. Andrews,et al.  Stress-Rupture Life of Kevlar 49/Epoxy Spherical Pressure Vessels , 1985 .

[23]  P. Rousseeuw.  Least Median of Squares Regression , 1984 .

[24]  Jin Zhang,et al.  A New and Efficient Estimation Method for the Generalized Pareto Distribution , 2009, Technometrics.

[25]  P. J. Huber.  Robust Regression: Asymptotics, Conjectures and Monte Carlo , 1973 .

[26]  M. Naghettini.  Parameter and Quantile Estimation , 2017 .

[27]  Vartan Choulakian,et al.  Goodness-of-Fit Tests for the Generalized Pareto Distribution , 2001, Technometrics.

[28]  D. Ferger.  Minimum distance estimation in normed linear spaces with Donsker-classes , 2010 .

[29]  A. Hadi,et al.  Fitting the Generalized Pareto Distribution to Data , 1997 .

[30]  E. Castillo.  Extreme value and related models with applications in engineering and science , 2005 .

[31]  J. Einmahl.  On the standarized empirical process. , 1989 .

[32]  Richard L. Smith,et al.  Models for exceedances over high thresholds , 1990 .

[33]  Michael L. Feldstein,et al.  The generalized Pareto law as a model for progressively censored survival data , 1979 .

[34]  Dimitris N. Politis,et al.  Model-free model-fitting and predictive distributions , 2010 .

[35]  S. Cabras,et al.  A default Bayesian procedure for the generalized Pareto distribution , 2007 .

[36]  Kunal N. Chaudhury,et al.  On the Convergence of the IRLS Algorithm in Non-Local Patch Regression , 2013, IEEE Signal Processing Letters.

[37]  Eric P. Smith,et al.  An Introduction to Statistical Modeling of Extreme Values , 2002, Technometrics.

[38]  S. Kotz,et al.  Parameter estimation of the generalized Pareto distribution—Part II , 2010 .

[39]  Seongjoo Song,et al.  A quantile estimation for massive data with generalized Pareto distribution , 2012, Comput. Stat. Data Anal..

[40]  Isabel Serra,et al.  Likelihood inference for generalized Pareto distribution , 2015, Comput. Stat. Data Anal..

[41]  E. J. Gumbel,et al.  Statistics of Extremes. , 1960 .

[42]  A. Keyhani,et al.  Iteratively reweighted least squares for maximum likelihood identification of synchronous machine parameters from on-line tests , 1999 .

[43]  W. R. Schucany,et al.  Minimum Distance and Robust Estimation , 1980 .

[44]  L. Haan,et al.  Extreme value theory : an introduction , 2006 .

[45]  On the use of partial probability weighted moments in the analysis of hydrological extremes , 2007 .

[46]  A. V. D. Vaart,et al.  Asymptotic Statistics: Frontmatter , 1998 .

[47]  Jin Zhang.  Improving on Estimation for the Generalized Pareto Distribution , 2010, Technometrics.