The skew generalized t distribution as the scale mixture of a skew exponential power distribution and its applications in robust estimation

In this paper, we consider the family of skew generalized t (SGT) distributions originally introduced by Theodossiou [P. Theodossiou, Financial data and the skewed generalized t distribution, Manage. Sci. Part 1 44 (12) ( 1998), pp. 1650–1661] as a skew extension of the generalized t (GT) distribution. The SGT distribution family warrants special attention, because it encompasses distributions having both heavy tails and skewness, and many of the widely used distributions such as Student's t, normal, Hansen's skew t, exponential power, and skew exponential power (SEP) distributions are included as limiting or special cases in the SGT family. We show that the SGT distribution can be obtained as the scale mixture of the SEP and generalized gamma distributions. We investigate several properties of the SGT distribution and consider the maximum likelihood estimation of the location, scale, and skewness parameters under the assumption that the shape parameters are known. We show that if the shape parameters are estimated along with the location, scale, and skewness parameters, the influence function for the maximum likelihood estimators becomes unbounded. We obtain the necessary conditions to ensure the uniqueness of the maximum likelihood estimators for the location, scale, and skewness parameters, with known shape parameters. We provide a simple iterative re-weighting algorithm to compute the maximum likelihood estimates for the location, scale, and skewness parameters and show that this simple algorithm can be identified as an EM-type algorithm. We finally present two applications of the SGT distributions in robust estimation.

[1]  B. Hansen.  Autoregressive Conditional Density Estimation , 1994 .

[2]  Richard L. Smith,et al.  A Comparison of Maximum Likelihood and Bayesian Estimators for the Three‐Parameter Weibull Distribution , 1987 .

[3]  G. S. Mudholkar,et al.  The epsilon-skew-normal distribution for analyzing near-normal data , 2000 .

[4]  Peter J. Rousseeuw,et al.  Robust Regression and Outlier Detection , 2005, Wiley Series in Probability and Statistics.

[5]  M. Genton,et al.  Flexible Class of Skew‐Symmetric Distributions , 2004 .

[6]  Joseph Lipka,et al.  A Table of Integrals , 2010 .

[7]  Anna Clara Monti,et al.  Inferential Aspects of the Skew Exponential Power Distribution , 2004 .

[8]  F. Chang,et al.  Some skew-symmetric models , 2002 .

[9]  Richard J. Butler,et al.  Some Generalized Mixture Distributions with an Application to Unemployment Duration , 1987 .

[10]  A note on rescalings, reparametrizations and classes of distributions , 2006 .

[11]  N. L. Johnson,et al.  Continuous Univariate Distributions. , 1995 .

[12]  S. Nadarajah,et al.  Skewed distributions generated by the normal kernel , 2003 .

[13]  S. Nadarajah,et al.  Skew Models II , 2007 .

[14]  Turan G. Bali,et al.  A conditional-SGT-VaR approach with alternative GARCH models , 2007, Ann. Oper. Res..

[15]  A. Azzalini.  A class of distributions which includes the normal ones , 1985 .

[16]  M. Steel,et al.  On Bayesian Modelling of Fat Tails and Skewness , 1998 .

[17]  W. N. Street,et al.  Financial Data and the Skewed Generalized T Distribution , 1998 .

[18]  G. Styan,et al.  On the Existence and Uniqueness of the Maximum Likelihood Estimate of a Vector-Valued Parameter in Fixed-Size Samples , 1981 .

[19]  D. Rubin,et al.  ML ESTIMATION OF THE t DISTRIBUTION USING EM AND ITS EXTENSIONS, ECM AND ECME , 1999 .

[20]  K. Song.  Rényi information, loglikelihood and an intrinsic distribution measure , 2001 .

[21]  Charles R. Johnson,et al.  Matrix analysis , 1985, Statistical Inference for Engineers and Data Scientists.

[22]  Richard J. Butler,et al.  Robust and Partially Adaptive Estimation of Regression Models , 1990 .

[23]  I. S. Gradshteyn,et al.  Table of Integrals, Series, and Products , 1976 .

[24]  G. C. Tiao,et al.  A Further Look at Robustness via Bayes's Theorem , 1962 .

[25]  Jeremy MG Taylor,et al.  Robust Statistical Modeling Using the t Distribution , 1989 .

[26]  M. C. Jones,et al.  A skew extension of the t‐distribution, with applications , 2003 .

[27]  Jose A. Romagnoli,et al.  Generalized T distribution and its applications to process data reconciliation and process monitoring , 2005 .

[28]  J. A. Romagnoli,et al.  A Framework for Robust Data Reconciliation Based on a Generalized Objective Function , 2003 .

[29]  A. Azzalini,et al.  Distributions generated by perturbation of symmetry with emphasis on a multivariate skew t‐distribution , 2003, 0911.2342.

[30]  Richard D. F. Harris,et al.  The Empirical Distribution of UK and US Stock Returns , 2001 .

[31]  Chris Jones,et al.  A skew t distribution , 2000 .

[32]  O. Arslan,et al.  Robust Location and Scale Estimation Based on the Univariate Generalized t (GT) Distribution , 2003 .

[33]  M. Steel,et al.  Modeling and Inference with υ-Spherical Distributions , 1995 .

[34]  Francis Galton F.R.S..  IV. Statistics by intercomparison, with remarks on the law of frequency of error , 1875 .

[35]  D. Ruppert.  Robust Statistics: The Approach Based on Influence Functions , 1987 .

[36]  Turan G. Bali,et al.  The role of autoregressive conditional skewness and kurtosis in the estimation of conditional VaR , 2008 .

[37]  Peter J. Rousseeuw,et al.  Robust regression and outlier detection , 1987 .

[38]  James V. Hansen,et al.  Partially adaptive robust estimation of regression models and applications , 2006, Eur. J. Oper. Res..

[39]  D. Rubin,et al.  The ECME algorithm: A simple extension of EM and ECM with faster monotone convergence , 1994 .

[40]  Vance L. Martin,et al.  Robust Estimation, Nonnormalities, and Generalized Exponential Distributions , 1993 .

[41]  Andre Lucas,et al.  Robustness of the student t based M-estimator , 1997 .

[42]  Julie Zhou.  Robust Estimationに , 2009 .

[43]  James B. McDonald,et al.  Partially Adaptive Estimation of Regression Models via the Generalized T Distribution , 1988, Econometric Theory.