Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/49359 
Year of Publication: 
2004
Series/Report no.: 
Technical Report No. 2004,18
Publisher: 
Universität Dortmund, Sonderforschungsbereich 475 - Komplexitätsreduktion in Multivariaten Datenstrukturen, Dortmund
Abstract: 
In this note we present a direct and simple approach to obtain bounds on the asymptotic minimax risk for the estimation of restrained binominal and multinominal proportions. Quadratic, normalized quadratic and entropy loss are considered and it is demonstrated that in all cases linear estimators are asymptotically minimax optimal. For the quadratic loss function the asymptotic minimax rsik does not change unless a neighborhood of the point 1/2 is excluded by the restrictions on the parameter space. For the two other loss functions the asymptotic minimax risks remain unchanged if additional knowledge about the location of the unknown probability of success is imposed. The results are also extended to the problem of minimax estimation of a vector of contrained multinominal propabilities.
Subjects: 
binominal distribution
multinominal distribution
entropy loss
quadratic loss
constrained parameter space
least favourable distribution
Document Type: 
Working Paper

Files in This Item:
File
Size
212.44 kB
221.86 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.