The problems arising when there are outliers in a data set that follow a circular distribution are considered. A robust estimation of the unknown parameters is obtained using the methods of weighted likelihood and minimum disparity, each of which is defined for a general parametric family of circular data. The class of power divergence and the related residual adjustment function is investigated in order to improve the performance of the two methods which are studied for the Von Mises (circular normal) and the Wrapped Normal distributions. The techniques are illustrated via two examples based on a real data set and a Monte Carlo study, which also enables the discussion of various computational aspects.

Robust Estimation in Circular Data

AGOSTINELLI, Claudio
2007-01-01

Abstract

The problems arising when there are outliers in a data set that follow a circular distribution are considered. A robust estimation of the unknown parameters is obtained using the methods of weighted likelihood and minimum disparity, each of which is defined for a general parametric family of circular data. The class of power divergence and the related residual adjustment function is investigated in order to improve the performance of the two methods which are studied for the Von Mises (circular normal) and the Wrapped Normal distributions. The techniques are illustrated via two examples based on a real data set and a Monte Carlo study, which also enables the discussion of various computational aspects.
File in questo prodotto:
File Dimensione Formato  
Agostinelli - 2007 - Robust estimation for circular data.pdf

non disponibili

Tipologia: Documento in Post-print
Licenza: Accesso chiuso-personale
Dimensione 202.79 kB
Formato Adobe PDF
202.79 kB Adobe PDF   Visualizza/Apri

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10278/25792
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact