A smoothing principle for the Huber and other location M-estimators

A smoothing principle for M-estimators is proposed. The smoothing depends on the sample size so that the resulting smoothed M-estimator coincides with the initial M-estimator when n->~. The smoothing principle is motivated by an analysis of the requirements in the proof of the Cramer-Rao bound. The principle can be applied to every M-estimator. A simulation study is carried out where smoothed Huber, ML-, and Bisquare M-estimators are compared with their non-smoothed counterparts and with Pitman estimators on data generated from several distributions with and without estimated scale. This leads to encouraging results for the smoothed estimators, and particularly the smoothed Huber estimator, as they improve upon the initial M-estimators particularly in the tail areas of the distributions of the estimators. The results are backed up by small sample asymptotics.

[1]  Stephan Morgenthaler,et al.  Point estimation: Location-and-scale configurations , 1991 .

[2]  E. Pitman,et al.  THE ESTIMATION OF THE LOCATION AND SCALE PARAMETERS OF A CONTINUOUS POPULATION OF ANY GIVEN FORM , 1939 .

[3]  David E. Tyler,et al.  On the finite sample breakdown points of redescending M-estimates of location , 2004 .

[4]  D. Berry,et al.  Statistics: Theory and Methods , 1990 .

[5]  J. Tukey.  Configural polysampling , 1987 .

[6]  C. J. Lawrence.  Robust estimates of location : survey and advances , 1975 .

[7]  Edward L. Ionides,et al.  MAXIMUM SMOOTHED LIKELIHOOD ESTIMATION , 2005 .

[8]  O. E. Barndorff-Nielsen,et al.  Infereni on full or partial parameters based on the standardized signed log likelihood ratio , 1986 .

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

[10]  O. Barndorff-Nielsen.  On a formula for the distribution of the maximum likelihood estimator , 1983 .

[11]  D. F. Andrews,et al.  Robust Estimates of Location: Survey and Advances. , 1975 .

[12]  Bruce G. Lindsay,et al.  A computational strategy for doubly smoothed MLE exemplified in the normal mixture model , 2010, Comput. Stat. Data Anal..

[13]  Laura Ventura.  Higher‐order approximations for pitman estimators and for optimal compromise estimators , 1998 .

[14]  Peter J. Rousseeuw,et al.  Robust estimation in very small samples , 2002 .

[15]  Frank R. Hampel,et al.  On the Philosophical Foundations of Statistics: Bridges to Huber’s work, and recent results , 1996 .

[16]  Frederick R. Forst,et al.  On robust estimation of the location parameter , 1980 .

[17]  Michael A. Fligner.  Small Sample Asymptotics , 1988 .

[18]  P. J. Huber,et al.  Robust statistics, data analysis, and computer intensive methods : in honor of Peter Huber's 60th birthday , 1996 .

[19]  H. Daniels.  The Asymptotic Efficiency of a Maximum Likelihood Estimator , 1961 .

[20]  C. Field,et al.  Small-sample asymptotic distributions of M-estimators of location , 1982 .

[21]  V. Yohai,et al.  Robust Statistics: Theory and Methods , 2006 .

[22]  May Thu Naing,et al.  Improved estimators for the location of double exponential distribution , 1989 .

[23]  Gabriela V. Cohen Freue,et al.  The Pitman estimator of the Cauchy location parameter , 2007 .

[24]  Peter Lischer.  Robust Statistical Methods in Interlaboratory Analytical Studies , 1996 .

[25]  H. Daniels.  Saddlepoint approximations for estimating equations , 1983 .

[26]  S. Rice,et al.  Saddle point approximation for the distribution of the sum of independent random variables , 1980, Advances in Applied Probability.

[27]  B. Ripley,et al.  Robust Statistics , 2018, Encyclopedia of Mathematical Geosciences.