Krzysztof Podgórski
Prefekt Statistiska institutionen, Professor
Maximizing leave-one-out likelihood for the location parameter of unbounded densities
Författare
Summary, in English
We propose simple estimation of the location parameter for a density that is unbounded at the mode. The estimator maximizes a modified likelihood in which the singular term in the full likelihood is left out, whenever the parameter value approaches a neighborhood of the singularity location. The consistency and super-efficiency of this maximum leave-one-out likelihood estimator is shown through a direct argument. The importance for estimation within parametric families is discussed and illustrated by an example involving the gamma mixture of normal distributions.
Avdelning/ar
- Statistiska institutionen
- Matematisk statistik
Publiceringsår
2015
Språk
Engelska
Sidor
19-38
Publikation/Tidskrift/Serie
Annals of the Institute of Statistical Mathematics
Volym
67
Issue
1
Dokumenttyp
Artikel i tidskrift
Förlag
Springer
Ämne
- Probability Theory and Statistics
Nyckelord
- Unbounded likelihood
- Location parameter
- Super-efficiency
- Generalized
- asymmetric Laplace distribution
Aktiv
Published
ISBN/ISSN/Övrigt
- ISSN: 0020-3157