Robust estimators of the generalized log-gamma distribution

We propose robust estimators of the generalized log-gamma distribution and, more generally, of location-shape-scale families of distributions. A (weighted) Qτ estimator minimizes a τ scale of the differences between empirical and theoretical quantiles. It is n 1/2 consistent; unfortunately, it is no...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Agostinelli, C., Marazzi, A., Yohai, V.J.
Formato: JOUR
Materias:
τ Estimators
Minimum quantile distance estimators
Weighted likelihood estimators
Acceso en línea:http://hdl.handle.net/20.500.12110/paper_00401706_v56_n1_p92_Agostinelli
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
id todo:paper_00401706_v56_n1_p92_Agostinelli
record_format dspace
spelling todo:paper_00401706_v56_n1_p92_Agostinelli2023-10-03T14:50:06Z Robust estimators of the generalized log-gamma distribution Agostinelli, C. Marazzi, A. Yohai, V.J. τ Estimators Minimum quantile distance estimators Weighted likelihood estimators We propose robust estimators of the generalized log-gamma distribution and, more generally, of location-shape-scale families of distributions. A (weighted) Qτ estimator minimizes a τ scale of the differences between empirical and theoretical quantiles. It is n 1/2 consistent; unfortunately, it is not asymptotically normal and, therefore, inconvenient for inference. However, it is a convenient starting point for a one-step weighted likelihood estimator, where the weights are based on a disparity measure between the model density and a kernel density estimate. The one-step weighted likelihood estimator is asymptotically normal and fully efficient under the model. It is also highly robust under outlier contamination. Supplementary materials are available online. © 2014 American Statistical Association and the American Society for Quality TECHNOMETRICS. JOUR info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by/2.5/ar http://hdl.handle.net/20.500.12110/paper_00401706_v56_n1_p92_Agostinelli
institution Universidad de Buenos Aires
institution_str I-28
repository_str R-134
collection Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic τ Estimators
Minimum quantile distance estimators
Weighted likelihood estimators
spellingShingle τ Estimators
Minimum quantile distance estimators
Weighted likelihood estimators
Agostinelli, C.
Marazzi, A.
Yohai, V.J.
Robust estimators of the generalized log-gamma distribution
topic_facet τ Estimators
Minimum quantile distance estimators
Weighted likelihood estimators
description We propose robust estimators of the generalized log-gamma distribution and, more generally, of location-shape-scale families of distributions. A (weighted) Qτ estimator minimizes a τ scale of the differences between empirical and theoretical quantiles. It is n 1/2 consistent; unfortunately, it is not asymptotically normal and, therefore, inconvenient for inference. However, it is a convenient starting point for a one-step weighted likelihood estimator, where the weights are based on a disparity measure between the model density and a kernel density estimate. The one-step weighted likelihood estimator is asymptotically normal and fully efficient under the model. It is also highly robust under outlier contamination. Supplementary materials are available online. © 2014 American Statistical Association and the American Society for Quality TECHNOMETRICS.
format JOUR
author Agostinelli, C.
Marazzi, A.
Yohai, V.J.
author_facet Agostinelli, C.
Marazzi, A.
Yohai, V.J.
author_sort Agostinelli, C.
title Robust estimators of the generalized log-gamma distribution
title_short Robust estimators of the generalized log-gamma distribution
title_full Robust estimators of the generalized log-gamma distribution
title_fullStr Robust estimators of the generalized log-gamma distribution
title_full_unstemmed Robust estimators of the generalized log-gamma distribution
title_sort robust estimators of the generalized log-gamma distribution
url http://hdl.handle.net/20.500.12110/paper_00401706_v56_n1_p92_Agostinelli
work_keys_str_mv AT agostinellic robustestimatorsofthegeneralizedloggammadistribution
AT marazzia robustestimatorsofthegeneralizedloggammadistribution
AT yohaivj robustestimatorsofthegeneralizedloggammadistribution
_version_ 1782029433529434112

Ejemplares similares