Premium
Reweighted loglikelihood method in robust estimation in a mixed unbalanced model
Author(s) -
Krug B.
Publication year - 1998
Publication title -
statistica neerlandica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 39
eISSN - 1467-9574
pISSN - 0039-0402
DOI - 10.1111/1467-9574.00077
Subject(s) - estimator , convergence (economics) , computer science , robustness (evolution) , scale (ratio) , algorithm , mathematical optimization , mathematics , statistics , quantum mechanics , economics , gene , economic growth , biochemistry , chemistry , physics
In the paper we introduce a method of computing M‐estimators in a mixed unbalanced model. The algorithm is based on weighted loglikelihood. It simultaneously finds the estimator of location and scale. Convergence of the algorithm to a solution of M‐equations is proved and a simulation study is given.