On the impact of robust statistics on imprecise probability models: A review

Robust statistics is concerned with statistical methods that still lead to reliable conclusions if an ideal model is only approximately true. More recently, the theory of imprecise probabilities was developed as a general methodology to model non-stochastic uncertainty (ambiguity) adequately, and has been successfully applied to many engineering problems. In robust statistics, small deviations from ideal models are modeled by certain neighborhoods. Since nearly all commonly used neighborhoods are imprecise probabilities, a large part of robust statistics can be seen as a special case of imprecise probabilities. Therefore, it seems quite promising to address problems in the theory of imprecise probabilities by trying to generalize results of robust statistics. In this review paper, we present some cases where this has already been done successfully and where the connections between (frequentist) robust statistics and imprecise probabilities are most striking.

[1]  Bjørnar Tessem,et al.  Interval probability propagation , 1992, Int. J. Approx. Reason..

[2]  Armando Mammino,et al.  Determination of parameters range in rock engineering by means of Random Set Theory , 2000, Reliab. Eng. Syst. Saf..

[3]  Michael Oberguggenberger,et al.  Imprecise probability methods for sensitivity analysis in engineering , 2007 .

[4]  Wolfgang Fellin,et al.  Analyzing uncertainty in civil engineering , 2005 .

[5]  Thomas Augustin.  Optimal decisions under complex uncertainty – basic notions and a general algorithm for data‐based decision making with partial prior knowledge described by interval probability , 2004 .

[6]  Lawrence C McCandless,et al.  A sensitivity analysis using information about measured confounders yielded improved uncertainty assessments for unmeasured confounding. , 2008, Journal of clinical epidemiology.

[7]  C. Manski.  Partial Identification of Probability Distributions , 2003 .

[8]  F. P. A. Coolen,et al.  Elementare Grundbegriffe einer Allgemeineren Wahrscheinlichkeitsrechnung, vol. I, Intervallwahrscheinlichkeit als Umfassendes Konzept , 2003 .

[9]  Charles F. Manski,et al.  Confidence Intervals for Partially Identified Parameters , 2003 .

[10]  Matthias Kohl,et al.  Numerical Contributions to the Asymptotic Theory of Robustness , 2005 .

[11]  Helmut Rieder.  A finite–sample minimax regression estimator , 1989 .

[12]  T. Hettmansperger,et al.  Simultaneous robust estimation of location and scale parameters: A minimum‐distance approach , 1998 .

[13]  R. Hable.  Minimum distance estimation in imprecise probability models , 2010 .

[14]  Scott Ferson,et al.  Constructing Probability Boxes and Dempster-Shafer Structures , 2003 .

[15]  T. Fine,et al.  Towards a Frequentist Theory of Upper and Lower Probability , 1982 .

[16]  Fabio Gagliardi Cozman,et al.  Partially Ordered Preferences in Decision Trees: Computing Strategies with Imprecision in Probabilities , 2005, IJCAI 2005.

[17]  Helmut Rieder.  Robustness in Structured Models , 1995 .

[18]  Robert Hable.  Data-based decisions under imprecise probability and least favorable models , 2009, Int. J. Approx. Reason..

[19]  Michael Oberguggenberger,et al.  Classical and imprecise probability methods for sensitivity analysis in engineering: A case study , 2009, Int. J. Approx. Reason..

[20]  Wolfgang Fellin,et al.  Reliability bounds through random sets , 2008 .

[21]  P. Hennequin,et al.  Quelques Aspects De La Statistique Robuste , 1981 .

[22]  Frank P. A. Coolen,et al.  Imprecise reliability: An introductory overview , 2007, Intelligence in Reliability Engineering.

[23]  P. Walley.  Statistical Reasoning with Imprecise Probabilities , 1990 .

[24]  Glenn Shafer,et al.  A Mathematical Theory of Evidence , 2020, A Mathematical Theory of Evidence.

[25]  Michael Beer,et al.  Comparison of uncertainty models in reliability analysis of offshore structures under marine corrosion , 2010 .

[26]  Dan Jackson,et al.  A Bound for Publication Bias Based on the Fraction of Unpublished Studies , 2004, Biometrics.

[27]  Thomas Augustin.  On the Suboptimality of the Generalized Bayes Rule and Robust Bayesian Procedures from the Decision Theoretic Point of View --- a Cautionary Note on Updating Imprecise Priors , 2003, ISIPTA.

[28]  Thomas Augustin,et al.  Decision making under incomplete data using the imprecise Dirichlet model , 2007, Int. J. Approx. Reason..

[29]  Sander Greenland,et al.  Monte Carlo sensitivity analysis and Bayesian analysis of smoking as an unmeasured confounder in a study of silica and lung cancer. , 2004, American journal of epidemiology.

[30]  V. Kreinovich,et al.  Experimental uncertainty estimation and statistics for data having interval uncertainty. , 2007 .

[31]  Marco Zaffalon,et al.  Conservative Inference Rule for Uncertain Reasoning under Incompleteness , 2009, J. Artif. Intell. Res..

[32]  Jim W. Hall,et al.  Uncertainty-based sensitivity indices for imprecise probability distributions , 2006, Reliab. Eng. Syst. Saf..

[33]  Peter Ruckdeschel,et al.  A motivation for $$1/ \sqrt{n}$$-shrinking neighborhoods , 2006 .

[34]  P. Millar.  Robust estimation via minimum distance methods , 1981 .

[35]  H. Rieder.  Least Favorable Pairs for Special Capacities , 1977 .

[36]  Geert Molenberghs,et al.  Incomplete Data in Clinical Studies: Analysis, Sensitivity, and Sensitivity Analysis , 2009 .

[37]  Robert Hafner.  Construction of Minimax-Tests for Bounded Families of Distribution- Functions , 1987 .

[38]  Gert de Cooman,et al.  ISIPTA'07: Proceedings of the Fifth International Symposium on Imprecise Probability: Theories and Applications , 2007 .

[39]  Gert de Cooman,et al.  Imprecise probability models for inference in exponential families , 2005, ISIPTA.

[40]  Michael Oberguggenberger.  Introductory remarks: Mathematical models of uncertainty , 2004 .

[41]  H. Rieder.  Robust asymptotic statistics , 1994 .

[42]  Didier Dubois,et al.  Unifying practical uncertainty representations - I: Generalized p-boxes , 2008, Int. J. Approx. Reason..

[43]  Gert de Cooman,et al.  Weak and strong laws of large numbers for coherent lower previsions , 2008 .

[44]  Peter Walley,et al.  Towards a unified theory of imprecise probability , 2000, Int. J. Approx. Reason..

[45]  Matthias C. M. Troffaes.  Decision making under uncertainty using imprecise probabilities , 2007, Int. J. Approx. Reason..

[46]  D. Donoho,et al.  The "Automatic" Robustness of Minimum Distance Functionals , 1988 .

[47]  Shinto Eguchi,et al.  Confidence Intervals and P‐Values for Meta‐Analysis with Publication Bias , 2007, Biometrics.

[48]  P. Walley.  Inferences from Multinomial Data: Learning About a Bag of Marbles , 1996 .

[49]  Th. Augustin,et al.  On Data-Based Checking of Hypotheses in the Presence of Uncertain Knowledge , 1999 .

[50]  Pauline Coolen-Schrijner,et al.  Imprecision in Statistical Theory and Practice , 2009 .

[51]  Bernhard Rüger,et al.  Grundlagen der Statistik und ihre Anwendungen : Festschrift für Kurt Weichselberger , 1995 .

[52]  Vladik Kreinovich,et al.  Exact Bounds on Finite Populations of Interval Data , 2005, Reliab. Comput..

[53]  P. Gustafson,et al.  Bayesian sensitivity analysis for unmeasured confounding in observational studies , 2007, Statistics in medicine.

[54]  V. Baumann,et al.  Eine parameterfreie Theorie der ungünstigsten Verteilungen für das Testen von Hypothesen , 1968 .

[55]  Brent Kreider,et al.  Partially Identifying Treatment Effects with an Application to Covering the Uninsured , 2009, The Journal of Human Resources.

[56]  Fabio Gagliardi Cozman.  Concentration Inequalities and Laws of Large Numbers under Epistemic Irrelevance , 2008, 0810.2821.

[57]  Gert de Cooman,et al.  Epistemic irrelevance in credal nets: The case of imprecise Markov trees , 2010, Int. J. Approx. Reason..

[58]  Damjan Skulj,et al.  Generalized Conditioning in Neighbourhood Models , 2005, ISIPTA.

[59]  Robert Hable.  A minimum distance estimator in an imprecise probability model - Computational aspects and applications , 2010, Int. J. Approx. Reason..

[60]  Francesca Molinari.  Partial identification of probability distributions with misclassified data , 2008 .

[61]  Gert de Cooman,et al.  Updating beliefs with incomplete observations , 2003, Artif. Intell..

[62]  Charles F. Manski,et al.  Partial identification with missing data: concepts and findings , 2005, Int. J. Approx. Reason..

[63]  Zhiwei Zhang,et al.  Likelihood-based confidence sets for partially identified parameters , 2009 .

[64]  Martin Browning,et al.  Best Nonparametric Bounds on Demand Responses , 2005 .

[65]  Arnold Neumaier,et al.  Potential Based Clouds in Robust Design Optimization , 2009 .

[66]  T. Bednarski.  On solutions of minimax test problems for special capacities , 1981 .

[67]  Damjan Skulj,et al.  Jeffrey's conditioning rule in neighbourhood models , 2006, Int. J. Approx. Reason..

[68]  Helmut Rieder,et al.  A Robust Asymptotic Testing Model , 1978 .

[69]  Tadesuz Bednarski.  On extension of the neyman-peabson lemma for capacities to dependent random variables , 1986 .

[70]  P. J. Huber.  Robust confidence limits , 1968 .

[71]  Andreas Buja,et al.  On the Huber-Strassen theorem , 1986 .

[72]  Robert Hable,et al.  Data-Based Decisions under Complex Uncertainty , 2009 .

[73]  P. J. Huber.  Robust Statistical Procedures , 1977 .

[74]  Geert Molenberghs,et al.  Ignorance and uncertainty regions as inferential tools in a sensitivity analysis , 2006 .

[75]  Thomas Augustin,et al.  Powerful algorithms for decision making under partial prior information and general ambiguity attitudes , 2005, ISIPTA.

[76]  Didier Dubois,et al.  Soft Methods for Handling Variability and Imprecision , 2008 .

[77]  Didier Dubois,et al.  The role of generalised p-boxes in imprecise probability models , 2009 .

[78]  Tadeusz Bednarski.  A robust asymptotic testing model for special capacities , 1985 .

[79]  P. Walley,et al.  A survey of concepts of independence for imprecise probabilities , 2000 .

[80]  Robert Hable.  Finite approximations of data-based decision problems under imprecise probabilities , 2009, Int. J. Approx. Reason..

[81]  J. Berger.  Robust Bayesian analysis : sensitivity to the prior , 1990 .

[82]  Jörg Stoye,et al.  More on Confidence Intervals for Partially Identified Parameters , 2008 .

[83]  Jörn Lembcke,et al.  The necessity of strongly subadditive capacities for Neyman-Pearson minimax tests , 1988 .

[84]  Thomas Augustin,et al.  Neyman–Pearson testing under interval probability by globally least favorable pairs: Reviewing Huber–Strassen theory and extending it to general interval probability , 2002 .

[85]  Brent Kreider,et al.  Bounding the effects of food insecurity on children's health outcomes. , 2009, Journal of health economics.

[86]  C. Manski,et al.  More on Monotone Instrumental Variables , 2009 .

[87]  Robert H Lyles,et al.  Inference for case-control studies when exposure status is both informatively missing and misclassified. , 2006, Statistics in medicine.

[88]  Tadeusz Bednarski,et al.  Binary Experiments, Minimax Tests and 2-Alternating Capacities , 1982 .

[89]  Dylan S. Small,et al.  Bounds on causal effects in three‐arm trials with non‐compliance , 2006 .

[90]  Monique Noirhomme-Fraiture,et al.  Symbolic Data Analysis and the SODAS Software , 2008 .

[91]  Anton Wallner,et al.  Bi-elastic Neighbourhood Models , 2003, ISIPTA.

[92]  Ross B. Corotis.  Risk communication with generalized uncertainty and linguistics , 2009 .

[93]  Patrick Brézillon,et al.  Lecture Notes in Artificial Intelligence , 1999 .

[94]  Geert Molenberghs,et al.  A multifaceted sensitivity analysis of the Slovenian public opinion survey data , 2009 .

[95]  Gm Gero Walter,et al.  Imprecision and Prior-data Conflict in Generalized Bayesian Inference , 2009 .

[96]  Frank R. Hampel,et al.  Nonadditive Probabilities in Statistics , 2009 .

[97]  F. Österreicher,et al.  On the construction of least favourable pairs of distributions , 1978 .

[98]  Scott Ferson,et al.  Sensitivity analysis using probability bounding , 2006, Reliab. Eng. Syst. Saf..

[99]  Charles F. Manski,et al.  Identification for Prediction and Decision , 2008 .

[100]  Gregory Levitin,et al.  Computational Intelligence in Reliability Engineering , 2007 .

[101]  Andreas Buja,et al.  Simultaneously least favorable experiments , 1984 .

[102]  Kurt Weichselberger.  The theory of interval-probability as a unifying concept for uncertainty , 2000, Int. J. Approx. Reason..

[103]  C. Paredis,et al.  The Value of Using Imprecise Probabilities in Engineering Design , 2006 .

[104]  Gm Gero Walter,et al.  Linear regression analysis under sets of conjugate priors , 2007 .

[105]  Jörg Stoye.  Partial Identification and Robust Treatment Choice: An Application to Young Offenders , 2009 .

[106]  Ulrich Pötter,et al.  Grundzüge der sozialwissenschaftlichen Statistik , 2001 .

[107]  Thomas Augustin,et al.  Lucenos' Discretization Method and its Application in Decision Making under Ambiguity , 2007 .

[108]  Robert Hafner,et al.  Simple construction of least favourable pairs of distributions and of robust tests for prokhorov-neighbourhoods , 1982 .

[109]  E. Moreno.  Global Bayesian Robustness for Some Classes of Prior Distributions , 2000 .

[110]  M. Beer,et al.  Engineering computation under uncertainty - Capabilities of non-traditional models , 2008 .

[111]  C. Huber,et al.  Theorie de la robustesse , 1986 .

[112]  Alexander Schied,et al.  Optimal Investments for Robust Utility Functionals in Complete Market Models , 2005, Math. Oper. Res..

[113]  Luis M. de Campos,et al.  Probability Intervals: a Tool for uncertain Reasoning , 1994, Int. J. Uncertain. Fuzziness Knowl. Based Syst..

[114]  P. J. Huber,et al.  Minimax Tests and the Neyman-Pearson Lemma for Capacities , 1973 .

[115]  Sander Greenland,et al.  Interval Estimation for Messy Observational Data , 2009, 1010.0306.

[116]  Thomas Augustin,et al.  Approximation of coherent lower probabilities by 2-monotone measures , 2009 .

[117]  Robert Hafner,et al.  Construction of minimax-tests for bounded families of probability-densities , 1993 .

[118]  Daniel Berleant,et al.  Representation and problem solving with Distribution Envelope Determination (DEnv) , 2004, Reliab. Eng. Syst. Saf..

[119]  Thomas Augustin.  On Decision Making under Ambiguous Prior and Sampling Information , 2001, ISIPTA.