Artículo

Estamos trabajando para incorporar este artículo al repositorio
Consulte el artículo en la página del editor
Consulte la política de Acceso Abierto del editor

Abstract:

Generalized Linear Models are routinely used in data analysis. Classical estimators are based on the maximum likelihood principle and it is well known that the presence of outliers can have a large impact on them. Several robust procedures have been presented in the literature, being redescending M-estimators the most widely accepted. Based on non-convex loss functions, these estimators need a robust initial estimate, which is often obtained by subsampling techniques. However, as the number of unknown parameters increases, the number of subsamples needed in order for this method to be robust, soon makes it infeasible. Furthermore the subsampling procedure provides a non deterministic starting point. A new method for computing a robust initial estimator is proposed. This method is deterministic and demands a relatively short computational time, even for large numbers of covariates. The proposed method is applied to M-estimators based on transformations. In addition, an iteratively reweighted least squares algorithm is proposed for the computation of the final estimates. The new methods are studied by means of Monte Carlo experiments. © 2018 Elsevier B.V.

Registro:

Documento: Artículo
Título:Initial robust estimation in generalized linear models
Autor:Agostinelli, C.; Valdora, M.; Yohai, V.J.
Filiación:Department of Mathematics, University of Trento, Trento, Italy
Departamento de Matematicas and Instituto de Cálculo, Facultad de Ciencias Exactas y Naturales, University of Buenos Aires, Argentina
Departamento de Matematicas and Institituto de Cálculo, Facultad de Ciencias Exactas y Naturales, University of Buenos Aires, CONICET, Argentina
Palabras clave:Initial estimates; Least squares estimators; M-estimators; Outliers; Poisson regression; Variance stabilizing transformations; Maximum likelihood estimation; Monte Carlo methods; Statistics; Initial estimate; Least-squares estimator; M-estimators; Outliers; Poisson regression; Iterative methods
Año:2019
Volumen:134
Página de inicio:144
Página de fin:156
DOI: http://dx.doi.org/10.1016/j.csda.2018.12.010
Título revista:Computational Statistics and Data Analysis
Título revista abreviado:Comput. Stat. Data Anal.
ISSN:01679473
CODEN:CSDAD
Registro:https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_01679473_v134_n_p144_Agostinelli

Referencias:

  • Agostinelli, C., Valdora, M., Yohai, V., (2018), https://cran.r-project.org/web/packages/poissonMT/index.html, Robust m-estimators based on transformations for poisson model, R package version 0.3-5; Alqallaf, F., Agostinelli, C., Robust inference in generalized linear models (2016) Comm. Statist. Simulation Comput., 45 (9), pp. 3053-3073
  • Bergesio, A., Yohai, V., Projection estimators for generalized linear models (2011) J. Amer. Statist. Assoc., 106, pp. 661-671
  • Bianco, A., Boente, G., Rodrigues, I., Resistant estimators in poisson and Gamma models with missing responses and an application to outlier detection (2013) J. Multivariate Anal., 114, pp. 209-226
  • Cantoni, E., Ronchetti, E., Robust inference for generalized linear models (2001) J. Amer. Statist. Assoc., 96, pp. 1022-1030
  • Connors, A., Speroff, T., Dawson, N., Thomas, C., Harrell, F., Wagner, D., Desbiens, N., Knaus, W., The effectiveness of right heart catheterization in the initial care of critically iii patients (1996) J. Am. Med. Assoc., 276 (11), pp. 889-897
  • Cook, R., Detection of influential observation in linear regression (1977) Technometrics, 19 (1), pp. 15-18
  • Dahl, D.B., (2016), https://CRAN.R-project.org/package=xtable, xtable: Export Tables to LaTeX or HTML, R package version 1.8-2; Künsch, H., Stefanski, L., Carroll, R., Conditionally unbiased bounded-influence estimation in general regression models, with applications to generalized linear models (1989) J. Amer. Statist. Assoc., 84, pp. 460-466
  • Lang, M., checkmate: Fast Argument Checks for Defensive R Programming (2017) R J., 9 (1), pp. 437-445. , https://journal.r-project.org/archive/2017/RJ-2017-028/index.html
  • Maechler, M., Rousseeuw, P., Croux, C., Todorov, V., Ruckstuhl, A., Salibian-Barrera, M., Verbeke, T., Anna di Palma, M., (2018), http://robustbase.r-forge.r-project.org/, robustbase: Basic Robust Statistics, R package version 0.93-1; Marazzi, A., (2018), https://CRAN.R-project.org/package=robcbi, bcbi: Conditionally Unbiased Bounded Influence Estimates, R package version 1.1-2; Marazzi, A., (2018), https://CRAN.R-project.org/package=robeth, robeth: R Functions for Robust Statistics, R package version 2.7-2; Maronna, R., Martin, R., Yohai, V., Robust Statistics. Theorey an Methods (2006), Wiley; McCullagh, P., Nelder, J., Generalized Linear Models (1989), second ed. Chapman and Hall/CRC; Peña, D., Yohai, V., A fast procedure for outlier diagnostics in large regression problems (1999) J. Amer. Statist. Assoc., 94, pp. 434-445
  • R Core Team, R: a language and environment for statistical computing (2018), https://www.R-project.org/, R Foundation for Statistical Computing Vienna, Austria; Rousseeuw, P., Leroy, A., Robust Regression and Outlier Detection (1987), Wiley and Sons; Tierney, L., Rossini, A.J., Li, N., Sevcikova, H., (2016), https://CRAN.R-project.org/package=snow, snow: Simple Network of Workstations, R package version 0.4-2; Valdora, M., Yohai, V., Robust estimators for generalized linear models (2014) J. Statist. Plann. Inference, 146, pp. 31-48
  • Venables, W.N., Ripley, B.D., Modern Applied Statistics with S (2002), http://www.stats.ox.ac.uk/pub/MASS4, fourth ed. Springer New York ISBN; Wang, J., Zamar, R., Marazzi, A., Yohai, V., Salibian-Barrera, M., Maronna, R., Zivot, E., Konis, K., (2017), https://CRAN.R-project.org/package=robust, robust: Port of the S+ ”Robust Library” R package version 0.4-18; Wei, T., Simko, V., (2017), https://github.com/taiyun/corrplot, R package ”corrplot”: visualization of a correlation matrix, (Version 0.84); Yu, H., Rmpi: parallel statistical computing in r (2002) R News, 2 (2), pp. 10-14. , https://cran.r-project.org/doc/Rnews/Rnews_2002-2.pdf

Citas:

---------- APA ----------
Agostinelli, C., Valdora, M. & Yohai, V.J. (2019) . Initial robust estimation in generalized linear models. Computational Statistics and Data Analysis, 134, 144-156.
http://dx.doi.org/10.1016/j.csda.2018.12.010
---------- CHICAGO ----------
Agostinelli, C., Valdora, M., Yohai, V.J. "Initial robust estimation in generalized linear models" . Computational Statistics and Data Analysis 134 (2019) : 144-156.
http://dx.doi.org/10.1016/j.csda.2018.12.010
---------- MLA ----------
Agostinelli, C., Valdora, M., Yohai, V.J. "Initial robust estimation in generalized linear models" . Computational Statistics and Data Analysis, vol. 134, 2019, pp. 144-156.
http://dx.doi.org/10.1016/j.csda.2018.12.010
---------- VANCOUVER ----------
Agostinelli, C., Valdora, M., Yohai, V.J. Initial robust estimation in generalized linear models. Comput. Stat. Data Anal. 2019;134:144-156.
http://dx.doi.org/10.1016/j.csda.2018.12.010