DOCSLIB.ORG

A Robust Statistics Approach to Minimum Variance Portfolio Optimization Liusha Yang∗, Romain Couillet†, Matthew R

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
A Robust Statistics Approach to Minimum Variance Portfolio Optimization Liusha Yang∗, Romain Couillet†, Matthew R 1 A Robust Statistics Approach to Minimum Variance Portfolio Optimization Liusha Yang∗, Romain Couillety, Matthew R. McKay∗ Abstract—We study the design of portfolios under a minimum In the finance literature, several approaches have been risk criterion. The performance of the optimized portfolio relies proposed to get around the problem of the scarcity of samples. on the accuracy of the estimated covariance matrix of the One approach is to impose some factor structure on the portfolio asset returns. For large portfolios, the number of available market returns is often of similar order to the number estimator of the covariance matrix [8, 9], which reduces the of assets, so that the sample covariance matrix performs poorly as number of parameters to be estimated. A second approach is a covariance estimator. Additionally, financial market data often to use as a covariance matrix estimator a weighted average of contain outliers which, if not correctly handled, may further the sample covariance matrix and another estimator, such as corrupt the covariance estimation. We address these shortcom- the 1-factor covariance matrix or the identity matrix [10, 11]. ings by studying the performance of a hybrid covariance matrix estimator based on Tyler’s robust M-estimator and on Ledoit- A third approach is a nonlinear shrinkage estimation approach Wolf’s shrinkage estimator while assuming samples with heavy- [12], which modifies each eigenvalue of the SCM under the tailed distribution. Employing recent results from random matrix framework of Markowitz’s portfolio selection. A fourth ap- theory, we develop a consistent estimator of (a scaled version proach comprises eigenvalue clipping methods [13–15] whose of) the realized portfolio risk, which is minimized by optimizing underlying idea is to ‘clean’ the SCM by filtering noisy online the shrinkage intensity. Our portfolio optimization method is shown via simulations to outperform existing methods both for eigenvalues claimed to convey little valuable information. This synthetic and real market data. approach has also been employed recently in proposing novel vaccine design strategies for infectious diseases [16, 17], and its theoretical foundations have been examined in [18]. A I. INTRODUCTION fifth method employs a bootstrap-corrected estimator for the optimal return and its asset allocation, which reduces the error The theory of portfolio optimization is generally associated of over-prediction of the in-sample return by bootstrapping with the classical mean-variance optimization framework of [6]. In contrast to all of these methods (which aim to improve Markowitz [1]. The pitfalls of the mean-variance analysis are the covariance matrix estimate), alternative methods have also mainly related to its sensitivity to the estimation error of been proposed which directly impose various constraints on the means and covariance matrix of the asset returns. It is the portfolio weights, such as a no-shortsale constraint [3], a nonetheless argued that estimates of the covariance matrix L norm constraint and a L norm constraint [19, 20]. By are more accurate than those of the expected returns [2, 3]. 1 2 bounding directly the portfolio-weight vector, it is demon- Thus, many studies concentrate on improving the performance strated that the estimation error can be reduced, particularly of the global minimum variance portfolio (GMVP), which when the portfolio size is large [19]. provides the lowest possible portfolio risk and involves only In addition to the problem of sample deficiency, it is often the covariance matrix estimate. the case that the return observations exhibit impulsiveness and The frequently used covariance estimator is the well-known local loss of stationarity [21], which is not addressed by the sample covariance matrix (SCM). However, covariance esti- methods mentioned above and leads to performance degrada- mates for portfolio optimization commonly involve few his- tion. The field of robust estimation [22–25] intends to deal with arXiv:1503.08013v1 [q-fin.PM] 27 Mar 2015 torical observations of sometimes up to a thousand assets. this problem. However, classical robust covariance estimators In such a case, the number of independent samples n may generally require n N and do not perform well (or are be small compared to the covariance matrix dimension N, not even defined) when n ' N, making them unsuitable which suggests a poor performance of the SCM. The impact for many modern applications. Recent works [26–32] based of the estimation error on the out-of-sample performance of on random matrix theory have therefore considered robust the GMVP based on the SCM has already been analyzed in estimation in the n ' N regime. Two hybrid robust shrinkage [4–7]. covariance matrix estimates have been proposed in parallel in ∗L. Yang and M. R. McKay are with the Department of [29, 30] and in [31], respectively, both of which estimators Electronic and Computer Engineering, Hong Kong University of are built upon Tyler’s robust M-estimatior [23] and Ledoit- Science and Technology, Clear Water Bay, Kowloon, Hong Kong. Wolf’s shrinkage approach [11]. In [32], the authors show, by (email:[email protected];[email protected]). yR. Couillet is with the Laboratoire de Signaux et Systmes (L2S, means of random matrix theory, that in the large n; N regime UMR8506), CNRS-CentraleSupelec-Universit´ e´ Paris-Sud, 3 rue Joliot-Curie, and under the assumption of elliptical vector observations, the 91192 Gif-sur-Yvette, France (email:[email protected]). estimators in [29, 30] and [31] perform essentially the same Yang and McKay’s work is supported by the Hong Kong Research Grants Council under grant number 16206914. Couillet’s work is supported by the and can be analyzed thanks to their asymptotic closeness to ERC MORE EC–120133. well-known random matrix models. Therefore, in this paper, 2 we concentrate on the estimator studied in [29, 30], which we in particular the class of elliptical distributions, including the denote by C^ ST (ST standing for shrinkage Tyler). Namely, for multivariate normal distribution, exponential distribution and N ^ independent samples x1; :::; xn 2 R with zero mean, CST the multivariate Student-T distribution as special cases. This is the unique solution to the fixed-point equation model for xt leads to tractable and adoptable design solutions n T and is a commonly used approximation of the impulsive nature 1 X xtxt C^ ST(ρ) = (1 − ρ) + ρIN of financial data [10]. n 1 T ^ −1 N t=1 N xt CST(ρ)xt Let h 2 R denote the portfolio selection, i.e., the vector of asset holdings in units of currency normalized by the total for any ρ 2 (maxf1 − N ; 0g; 1]. It should be noted that the n outstanding wealth, satisfying hT 1 = 1. In this paper, short- shrinkage structure even allows n < N. N selling is allowed, and thus the portfolio weights may be This paper designs a novel minimum variance portfolio op- negative. Then the portfolio variance (or risk) over the invest- timization strategy based on C^ with a risk-minimizing (in- ST ment period of interest is defined as σ2(h) = E[jhT x j2] = stead of Frobenius norm minimizing [32]) shrinkage parameter t hT C h [1]. Accordingly, the GMVP selection problem can ρ. We first characterize the out-of-sample risk of the minimum N be formulated as the following quadratic optimization problem variance portfolio with plug-in ST for all ρ within a specified with a linear constraint: range. This is done by analyzing the uniform convergence of 2 T the achieved realized risk on ρ in the double limit regime, min σ (h) s:t: h 1N = 1: h where N; n ! 1, with cN = N=n ! c 2 (0; 1). We subsequently provide a consistent estimator of the realized This has the well-known solution portfolio risk (or, more precisely, a scaled version of it) that −1 CN 1N is defined only in terms of the observed returns. Based on hGMVP = 1T C−11 this, we obtain a risk-optimized ST covariance estimator by N N N optimizing online over ρ, and thus our optimized portfolio. and the corresponding portfolio risk is The proposed portfolio selection is shown to achieve supe- 1 2 (2) rior performance over the competing methods in [11, 31–33] σ (hGMVP) = T −1 : 1 C 1N in minimizing the realized portfolio risk under the GMVP N N framework for impulsive data. The outperformance of our Here, (2) represents the theoretical minimum portfolio risk portfolio optimization strategy compared to other methods is bound, attained upon knowing the covariance matrix CN demonstrated through Monte Carlo simulations with ellipti- exactly. In practice, CN is unknown, and instead we form an ^ cally distributed samples, as well as with real data of historical estimate, denoted by CN . Thus, the GMVP selection based ^ (daily) stock returns from Hong Kong’s Hang Seng Index on the plug-in estimator CN is given by (HSI). ^ −1 ^ CN 1N Notations: Boldface upper case letters denote matrices, hGMVP = : 1T C^ −11 boldface lower case letters denote column vectors, and stan- N N N T dard lower case letters denote scalars. (·) denotes transpose. The quality of h^GMVP, implemented based on the in-sample IN denotes the N × N identity matrix and 1N denotes an N- covariance prediction C^ N , can be measured by its achieved dimensional vector with all entries equal to one. tr[·] denotes out-of-sample (or “realized”) portfolio risk: the matrix trace operator. and denote the real and complex R C T ^ −1 ^ −1 fields of dimension specified by a superscript. k·k denotes the 2 1N CN CN CN 1N σ h^GMVP = : T ^ −1 2 Euclidean norm for vectors and the spectral norm for matrices. (1N CN 1N ) The Dirac measure at point x is denoted by δ .
Load more

Recommended publications
  • Lecture 12 Robust Estimation
    Lecture 12 Robust Estimation Prof. Dr. Svetlozar Rachev Institute for Statistics and Mathematical Economics University of Karlsruhe Financial Econometrics, Summer Semester 2007 Prof. Dr. Svetlozar Rachev Institute for Statistics and MathematicalLecture Economics 12 Robust University Estimation of Karlsruhe Copyright These lecture-notes cannot be copied and/or distributed without permission. The material is based on the text-book: Financial Econometrics: From Basics to Advanced Modeling Techniques (Wiley-Finance, Frank J. Fabozzi Series) by Svetlozar T. Rachev, Stefan Mittnik, Frank Fabozzi, Sergio M. Focardi,Teo Jaˇsic`. Prof. Dr. Svetlozar Rachev Institute for Statistics and MathematicalLecture Economics 12 Robust University Estimation of Karlsruhe Outline I Robust statistics. I Robust estimators of regressions. I Illustration: robustness of the corporate bond yield spread model. Prof. Dr. Svetlozar Rachev Institute for Statistics and MathematicalLecture Economics 12 Robust University Estimation of Karlsruhe Robust Statistics I Robust statistics addresses the problem of making estimates that are insensitive to small changes in the basic assumptions of the statistical models employed. I The concepts and methods of robust statistics originated in the 1950s. However, the concepts of robust statistics had been used much earlier. I Robust statistics: 1. assesses the changes in estimates due to small changes in the basic assumptions; 2. creates new estimates that are insensitive to small changes in some of the assumptions. I Robust statistics is also useful to separate the contribution of the tails from the contribution of the body of the data. Prof. Dr. Svetlozar Rachev Institute for Statistics and MathematicalLecture Economics 12 Robust University Estimation of Karlsruhe Robust Statistics I Peter Huber observed, that robust, distribution-free, and nonparametrical actually are not closely related properties.
    [Show full text]
  • Robustness of Parametric and Nonparametric Tests Under Non-Normality for Two Independent Sample
    International Journal of Management and Applied Science, ISSN: 2394-7926 Volume-4, Issue-4, Apr.-2018 http://iraj.in ROBUSTNESS OF PARAMETRIC AND NONPARAMETRIC TESTS UNDER NON-NORMALITY FOR TWO INDEPENDENT SAMPLE 1USMAN, M., 2IBRAHIM, N 1Dept of Statistics, Fed. Polytechnic Bali, Taraba State, Dept of Maths and Statistcs, Fed. Polytechnic MubiNigeria E-mail: [email protected] Abstract - Robust statistical methods have been developed for many common problems, such as estimating location, scale and regression parameters. One motivation is to provide methods with good performance when there are small departures from parametric distributions. This study was aimed to investigates the performance oft-test, Mann Whitney U test and Kolmogorov Smirnov test procedures on independent samples from unrelated population, under situations where the basic assumptions of parametric are not met for different sample size. Testing hypothesis on equality of means require assumptions to be made about the format of the data to be employed. Sometimes the test may depend on the assumption that a sample comes from a distribution in a particular family; if there is a doubt, then a non-parametric tests like Mann Whitney U test orKolmogorov Smirnov test is employed. Random samples were simulated from Normal, Uniform, Exponential, Beta and Gamma distributions. The three tests procedures were applied on the simulated data sets at various sample sizes (small and moderate) and their Type I error and power of the test were studied in both situations under study. Keywords - Non-normal,Independent Sample, T-test,Mann Whitney U test and Kolmogorov Smirnov test. I. INTRODUCTION assumptions about your data, but it may also require the data to be an independent random sample4.
    [Show full text]
  • Robust Methods in Biostatistics
    Robust Methods in Biostatistics Stephane Heritier The George Institute for International Health, University of Sydney, Australia Eva Cantoni Department of Econometrics, University of Geneva, Switzerland Samuel Copt Merck Serono International, Geneva, Switzerland Maria-Pia Victoria-Feser HEC Section, University of Geneva, Switzerland A John Wiley and Sons, Ltd, Publication Robust Methods in Biostatistics WILEY SERIES IN PROBABILITY AND STATISTICS Established by WALTER A. SHEWHART and SAMUEL S. WILKS Editors David J. Balding, Noel A. C. Cressie, Garrett M. Fitzmaurice, Iain M. Johnstone, Geert Molenberghs, David W. Scott, Adrian F. M. Smith, Ruey S. Tsay, Sanford Weisberg, Harvey Goldstein. Editors Emeriti Vic Barnett, J. Stuart Hunter, Jozef L. Teugels A complete list of the titles in this series appears at the end of this volume. Robust Methods in Biostatistics Stephane Heritier The George Institute for International Health, University of Sydney, Australia Eva Cantoni Department of Econometrics, University of Geneva, Switzerland Samuel Copt Merck Serono International, Geneva, Switzerland Maria-Pia Victoria-Feser HEC Section, University of Geneva, Switzerland A John Wiley and Sons, Ltd, Publication This edition first published 2009 c 2009 John Wiley & Sons Ltd Registered office John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com. The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.
    [Show full text]
  • A Guide to Robust Statistical Methods in Neuroscience
    A GUIDE TO ROBUST STATISTICAL METHODS IN NEUROSCIENCE Authors: Rand R. Wilcox1∗, Guillaume A. Rousselet2 1. Dept. of Psychology, University of Southern California, Los Angeles, CA 90089-1061, USA 2. Institute of Neuroscience and Psychology, College of Medical, Veterinary and Life Sciences, University of Glasgow, 58 Hillhead Street, G12 8QB, Glasgow, UK ∗ Corresponding author: [email protected] ABSTRACT There is a vast array of new and improved methods for comparing groups and studying associations that offer the potential for substantially increasing power, providing improved control over the probability of a Type I error, and yielding a deeper and more nuanced understanding of data. These new techniques effectively deal with four insights into when and why conventional methods can be unsatisfactory. But for the non-statistician, the vast array of new and improved techniques for comparing groups and studying associations can seem daunting, simply because there are so many new methods that are now available. The paper briefly reviews when and why conventional methods can have relatively low power and yield misleading results. The main goal is to suggest some general guidelines regarding when, how and why certain modern techniques might be used. Keywords: Non-normality, heteroscedasticity, skewed distributions, outliers, curvature. 1 1 Introduction The typical introductory statistics course covers classic methods for comparing groups (e.g., Student's t-test, the ANOVA F test and the Wilcoxon{Mann{Whitney test) and studying associations (e.g., Pearson's correlation and least squares regression). The two-sample Stu- dent's t-test and the ANOVA F test assume that sampling is from normal distributions and that the population variances are identical, which is generally known as the homoscedastic- ity assumption.
    [Show full text]
  • Robust Statistical Methods in R Using the WRS2 Package
    JSS Journal of Statistical Software MMMMMM YYYY, Volume VV, Issue II. doi: 10.18637/jss.v000.i00 Robust Statistical Methods in R Using the WRS2 Package Patrick Mair Rand Wilcox Harvard University University of Southern California Abstract In this manuscript we present various robust statistical methods popular in the social sciences, and show how to apply them in R using the WRS2 package available on CRAN. We elaborate on robust location measures, and present robust t-test and ANOVA ver- sions for independent and dependent samples, including quantile ANOVA. Furthermore, we present on running interval smoothers as used in robust ANCOVA, strategies for com- paring discrete distributions, robust correlation measures and tests, and robust mediator models. Keywords: robust statistics, robust location measures, robust ANOVA, robust ANCOVA, robust mediation, robust correlation. 1. Introduction Data are rarely normal. Yet many classical approaches in inferential statistics assume nor- mally distributed data, especially when it comes to small samples. For large samples the central limit theorem basically tells us that we do not have to worry too much. Unfortu- nately, things are much more complex than that, especially in the case of prominent, \dan- gerous" normality deviations such as skewed distributions, data with outliers, or heavy-tailed distributions. Before elaborating on consequences of these violations within the context of statistical testing and estimation, let us look at the impact of normality deviations from a purely descriptive angle. It is trivial that the mean can be heavily affected by outliers or highly skewed distribu- tional shapes. Computing the mean on such data would not give us the \typical" participant; it is just not a good location measure to characterize the sample.
    [Show full text]
  • Robustbase: Basic Robust Statistics
    Package ‘robustbase’ June 2, 2021 Version 0.93-8 VersionNote Released 0.93-7 on 2021-01-04 to CRAN Date 2021-06-01 Title Basic Robust Statistics URL http://robustbase.r-forge.r-project.org/ Description ``Essential'' Robust Statistics. Tools allowing to analyze data with robust methods. This includes regression methodology including model selections and multivariate statistics where we strive to cover the book ``Robust Statistics, Theory and Methods'' by 'Maronna, Martin and Yohai'; Wiley 2006. Depends R (>= 3.5.0) Imports stats, graphics, utils, methods, DEoptimR Suggests grid, MASS, lattice, boot, cluster, Matrix, robust, fit.models, MPV, xtable, ggplot2, GGally, RColorBrewer, reshape2, sfsmisc, catdata, doParallel, foreach, skewt SuggestsNote mostly only because of vignette graphics and simulation Enhances robustX, rrcov, matrixStats, quantreg, Hmisc EnhancesNote linked to in man/*.Rd LazyData yes NeedsCompilation yes License GPL (>= 2) Author Martin Maechler [aut, cre] (<https://orcid.org/0000-0002-8685-9910>), Peter Rousseeuw [ctb] (Qn and Sn), Christophe Croux [ctb] (Qn and Sn), Valentin Todorov [aut] (most robust Cov), Andreas Ruckstuhl [aut] (nlrob, anova, glmrob), Matias Salibian-Barrera [aut] (lmrob orig.), Tobias Verbeke [ctb, fnd] (mc, adjbox), Manuel Koller [aut] (mc, lmrob, psi-func.), Eduardo L. T. Conceicao [aut] (MM-, tau-, CM-, and MTL- nlrob), Maria Anna di Palma [ctb] (initial version of Comedian) 1 2 R topics documented: Maintainer Martin Maechler <[email protected]> Repository CRAN Date/Publication 2021-06-02 10:20:02 UTC R topics documented: adjbox . .4 adjboxStats . .7 adjOutlyingness . .9 aircraft . 12 airmay . 13 alcohol . 14 ambientNOxCH . 15 Animals2 . 18 anova.glmrob . 19 anova.lmrob .
    [Show full text]
  • Robust Statistics in BD Facsdiva™ Software
    BD Biosciences Tech Note June 2012 Robust Statistics in BD FACSDiva™ Software Robust Statistics in BD FACSDiva™ Software Robust statistics defined Robust statistics provide an alternative approach to classical statistical estimators such as mean, standard deviation (SD), and percent coefficient of variation (%CV). These alternative procedures are more resistant to the statistical influences of outlying events in a sample population—hence the term “robust.” Real data sets often contain gross outliers, and it is impractical to systematically attempt to remove all outliers by gating procedures or other rule sets. The robust equivalent of the mean statistic is the median. The robust SD is designated rSD and the percent robust CV is designated %rCV. For perfectly normal distributions, classical and robust statistics give the same results. How robust statistics are calculated in BD FACSDiva™ software Median The mean, or average, is the sum of all the values divided by the number of values. For example, the mean of the values of [13, 10, 11, 12, 114] is 160 ÷ 5 = 32. If the outlier value of 114 is excluded, the mean is 11.5. The median is defined as the midpoint, or the 50th percentile of the values. It is the statistical center of the population. Half the values should be above the median and half below. In the previous example, the median is 12 (13 and 114 are above; 10 and 11 are below). Note that this is close to the mean with the outlier excluded. Robust standard deviation (rSD) The classical SD is a function of the deviation of individual data points to the mean of the population.
    [Show full text]
  • Application of Robust Regression and Bootstrap in Productivity Analysis of GERD Variable in EU27
    METHODOLOGY Application of Robust Regression and Bootstrap in Productivity Analysis of GERD Variable in EU27 Dagmar Blatná1 | University of Economics, Prague, Czech Republic Abstract Th e GERD is one of Europe 2020 headline indicators being tracked within the Europe 2020 strategy. Th e head- line indicator is the 3% target for the GERD to be reached within the EU by 2020. Eurostat defi nes “GERD” as total gross domestic expenditure on research and experimental development in a percentage of GDP. GERD depends on numerous factors of a general economic background, namely of employment, innovation and research, science and technology. Th e values of these indicators vary among the European countries, and consequently the occurrence of outliers can be anticipated in corresponding analyses. In such a case, a clas- sical statistical approach – the least squares method – can be highly unreliable, the robust regression methods representing an acceptable and useful tool. Th e aim of the present paper is to demonstrate the advantages of robust regression and applicability of the bootstrap approach in regression based on both classical and robust methods. Keywords JEL code LTS regression, MM regression, outliers, leverage points, bootstrap, GERD C19, C49, O11, C13 INTRODUCTION GERD represents total gross domestic expenditure on research and experimental development (R&D) as a percentage of GDP (Eurostat), R&D expenditure capacity being regarded as an important factor of the economic growth. GERD is one of Europe 2020 indicator sets used by the European Commission to monitor headline strategy targets for the next decade –A Strategy for Smart, Sustainable and Inclusive Growth (every country should invest 3% of GDP in R&D by 2020).
    [Show full text]
  • Robust Statistics
    A Brief Overview of Robust Statistics Olfa Nasraoui Department of Computer Engineering & Computer Science University of Louisville, olfa.nasraoui_AT_louisville.edu Robust Statistical Estimators Robust statistics have recently emerged as a family of theories and techniques for estimating the parameters of a parametric model while dealing with deviations from idealized assumptions [Goo83,Hub81,HRRS86,RL87]. Examples of deviations include the contamination of data by gross errors, rounding and grouping errors, and departure from an assumed sample distribution. Gross errors or outliers are data severely deviating from the pattern set by the majority of the data. This type of error usually occurs due to mistakes in copying or computation. They can also be due to part of the data not fitting the same model, as in the case of data with multiple clusters. Gross errors are often the most dangerous type of errors. In fact, a single outlier can completely spoil the Least Squares estimate, causing it to break down. Rounding and grouping errors result from the inherent inaccuracy in the collection and recording of data which is usually rounded, grouped, or even coarsely classified. The departure from an assumed model means that real data can deviate from the often assumed normal distribution. The departure from the normal distribution can manifest itself in many ways such as in the form of skewed (asymmetric) or longer tailed distributions. M-estimators The ordinary Least Squares (LS) method to estimate parameters is not robust because its objective function, , increases indefinitely with the residuals between the data point and the estimated fit, with being the total number of data points in a data set.
    [Show full text]
  • Robust Statistics: a Brief Introduction and Overview
    Robust statistics: A brief introduction and overview by Frank Hampel E-Mail: [email protected] Research Report No. 94 March 2001 Seminar f¨urStatistik Eidgen¨ossische Technische Hochschule (ETH) CH-8092 Z¨urich Switzerland Invited talk in the Symposium “Robust Statistics and Fuzzy Techniques in Geodesy and GIS” held in ETH Zurich, March 12-16, 2001. Robust statistics: a brief introduction and overview Frank Hampel Seminar for Statistics, ETH Zurich, Switzerland E-Mail: [email protected] Abstract The paper gives a highly condensed first introduction into robust statistics and some guidance for the interpretation of the literature, with some consideration for the uses in geodetics in the background. Robust statistics is the stability theory of statistical procedures. It systematically investigates the effects of deviations from modelling assumptions on known procedures and, if necessary, develops new, better procedures. Common modelling assumptions are those of normality and of independence of the ran- dom errors. For both exist sophisticated theories with important practical consequences, and specifically normality can be replaced by any other reasonable parametric model (cf. Huber 1981, Hampel et al. 1986, and also Beran 1994). As a simple example, let us consider n “independent” measurements of the same quan- tity, for example a distance. In general they will differ, namely by what we now call the “random error”, and the question arises which value we should take as the “best estimate” of the unknown “true value”. This question was already considered by Gauss (1821; cf. also Huber 1972), and he, noticing that he needed the unknown error distribution to an- swer this question, turned the problem upside down and asked for that error distribution which made a rule “generally accepted as a good one”, namely the arithmetic mean, op- timal (in location or shift models).
    [Show full text]
  • Robust Weighted Likelihood Estimation of Exponential Parameters Ejaz S
    IEEE TRANSACTIONS ON RELIABILITY, VOL. 54, NO. 3, SEPTEMBER 2005 389 Robust Weighted Likelihood Estimation of Exponential Parameters Ejaz S. Ahmed, Andrei I. Volodin, and Abdulkadir. A. Hussein Abstract—The problem of estimating the parameter of an ex- • sample mean ponential distribution when a proportion of the observations are • -trimmed sample mean outliers is quite important to reliability applications. The method • quadratic risk of a statistics of weighted likelihood is applied to this problem, and a robust estimator of the exponential parameter is proposed. Interestingly, • , Convergence as fast as a, convergence faster the proposed estimator is an -trimmed mean type estimator. than The large-sample robustness properties of the new estimator are examined. Further, a Monte Carlo simulation study is conducted I. INTRODUCTION showing that the proposed estimator is, under a wide range of contaminated exponential models, more efficient than the usual HE EXPONENTIAL distribution is the most commonly maximum likelihood estimator in the sense of having a smaller T used model in reliability & life-testing analysis [1], [2]. risk, a measure combining bias & variability. An application of the We propose a method of weighted likelihood for robust estima- method to a data set on the failure times of throttles is presented. tion of the exponential distribution parameter. We establish that Index Terms—Exponential distribution, maximum likelihood es- the suggested weighted likelihood method provides -trimmed timation, robust estimation, weighted likelihood method. mean type estimators. Further, we investigate the robustness properties of the weighted likelihood estimator (WLE) in com- ACRONYMS1 parison with the usual maximum likelihood estimator (MLE). This can be viewed as an application of the classical likelihood • a.s.
    [Show full text]
  • Robust Improper Maximum Likelihood: Tuning, Computation, and a Comparison with Other Methods for Robust Gaussian Clustering
    ROBUST IMPROPER MAXIMUM LIKELIHOOD: TUNING, COMPUTATION, AND A COMPARISON WITH OTHER METHODS FOR ROBUST GAUSSIAN CLUSTERING Pietro Coretto∗ Christian Hennig† University of Salerno, Italy University College London, UK e-mail: [email protected] e-mail: [email protected] This is a preprint. The revised version of this paper is published as Coretto, P. and Hennig, C. (2016). Robust Improper Maximum Likelihood: Tuning, Computation, and a Comparison with Other Methods for Robust Gaussian Clustering. Journal of the American Statistical Association 111(516), pp. 1648–1659. [link] Abstract. The two main topics of this paper are the introduction of the “optimally tuned improper maximum likelihood estimator” (OTRIMLE) for robust clustering based on the multivariate Gaussian model for clusters, and a comprehensive simula- tion study comparing the OTRIMLE to Maximum Likelihood in Gaussian mixtures with and without noise component, mixtures of t-distributions, and the TCLUST ap- proach for trimmed clustering. The OTRIMLE uses an improper constant density for modelling outliers and noise. This can be chosen optimally so that the non-noise part of the data looks as close to a Gaussian mixture as possible. Some deviation from Gaussianity can be traded in for lowering the estimated noise proportion. Covariance matrix constraints and computation of the OTRIMLE are also treated. In the simula- tion study, all methods are confronted with setups in which their model assumptions are not exactly fulfilled, and in order to evaluate the experiments in a standardized way by misclassification rates, a new model-based definition of “true clusters” is intro- duced that deviates from the usual identification of mixture components with clusters.
    [Show full text]