On approximate validation of models:: a Kolmogorov–Smirnov-based approach

• E. del Barrio ^[1] ; H. Inouzhe ^[1] ; C. Matrán ^[1]
  1. [1] Universidad de Valladolid

     Universidad de Valladolid

     Valladolid, España

     
• Localización: Test: An Official Journal of the Spanish Society of Statistics and Operations Research, ISSN-e 1863-8260, ISSN 1133-0686, Vol. 29, Nº. 4, 2020, págs. 938-965
• Idioma: inglés
• DOI: 10.1007/s11749-019-00691-1
• Texto completo no disponible (Saber más ...)
• Resumen

  • Classical tests of fit typically reject a model for large enough real data samples. In contrast, often in statistical practice, a model offers a good description of the data even though it is not the ‘true’ random generator. We consider a more flexible approach based on contamination neighbourhoods: using trimming methods and the Kolmogorov metric, we introduce a functional statistic measuring departures from a contaminated model. We show how the plug-in estimator allows testing of fit for the (slightly) contaminated model vs sensible deviations from it, with uniformly exponentially small type I and type II error probabilities. We also address the asymptotic behaviour of the estimator showing that, under suitable regularity conditions, it asymptotically behaves as the supremum of a Gaussian process. As an application, we explore methods of comparison between descriptive models based on the paradigm of model falseness. We also include some connections of our approach with the false discovery rate setting, showing competitive behaviour when estimating the contamination level, and being applicable in a wider framework.

• Referencias bibliográficas
  • Álvarez-Esteban PC, del Barrio E, Cuesta-Albertos JA, Matrán C (2008) Trimmed comparison of distributions. J Am Stat Assoc 103:697–704
  • Álvarez-Esteban PC, del Barrio E, Cuesta-Albertos JA, Matrán C (2011) Uniqueness and approximate computation of optimal incomplete transportation...
  • Álvarez-Esteban PC, del Barrio E, Cuesta-Albertos JA, Matrán C (2012) Similarity of samples and trimming. Bernoulli 18:606–634
  • Álvarez-Esteban PC, del Barrio E, Cuesta-Albertos JA, Matrán C (2016) A contamination model for approximate stochastic order. Test 25:751–774
  • Barron A (1989) Uniformly powerful goodness of fit tests. Ann Stat 17:107–124
  • Berkson J (1938) Some difficulties of interpretation encountered in the application of the chi-square test. J Am Stat Assoc 33:526–536
  • Cárcamo J, Rodríguez L-A, Cuevas A (2019) Directional differentiability for supremum-type functionals: statistical applications. arXiv:1902.01136
  • Davies L (1995) Data features. Stat Neerl 49:185–245
  • Davies L (2014) Data analysis and approximate models, vol 133. Monographs on statistics and applied probability. CRC Press, Boca Raton
  • Davies L (2018) On p-values. Stat Sin 28(5):2823–2840
  • del Barrio E, Matrán C (2013) Rates of convergence for partial mass problems. Probab Theory Relat Fields 155:521–542
  • del Barrio E, Inouzhe H, Matrán C (2019) Box-constrained monotone L∞-approximations to Lipschitz regularizations, with applications to robust...
  • Donoho DL (1988) One sided inference about functionals of a density. Ann Stat 16:1390–1420
  • Genovese C, Wasserman L (2004) A stochastic process approach to false discovery control. Ann Stat 32(3):1035–1061
  • Hodges J Jr, Lehmann E (1954) Testing the approximate validity of statistical hypotheses. J R Stat Soc B 16(2):261–268
  • Huber PJ (1964) Robust estimation of a location parameter. Ann Math Stat 35:73–101
  • Lindsay B, Liu J (2009) Model assessment tools for a model false world. Stat Sci 24:303–318
  • Liu J, Lindsay B (2009) Building and using semiparametric tolerance regions for parametric multinomial models. Ann Stat 37:3644–3659
  • Massart P (1990) The tight constant in the Dvoretzky–Kiefer–Wolfowitz inequality. Ann Probab 18:1269–1283
  • Meinshausen N, Rice J (2006) Estimating the proportion of false null hypotheses among a large number of independently tested hypotheses. Ann...
  • Munk A, Czado C (1998) Nonparametric validation of similar distributions and assessment of goodness of fit. J R Stat Soc B 60:223–241
  • Owen AB (1995) Nonparametric likelihood confidence bands function for a distribution. J Am Stat Assoc 90(430):516–521
  • Raghavachari M (1973) Limiting distributions of Kolmogorov–Smirnov type statistics under the alternative. Ann Stat 1:67–73
  • Rieder H (1977) Least favorable pairs for special capacities. Ann Stat 5:909–921
  • Rieder H (1994) Robust asymptotic statistics. Springer, New York
  • Rudas T, Clogg CC, Lindsay BG (1994) A new index of fit based on mixture methods for the analysis of contingency tables. J R Stat Soc B 56(4):623–639
  • Shorack GR, Wellner JA (1986) Empirical processes with applications to statistics. Classics in applied mathematics. SIAM, Philadelphia