A multidimensional Newton-Raphson method and its applications to the existence of asymptotic Fn-estimators and their stochastic expansions
Given a suitable function Fn we define a class of estimators called asymptotic Fn-estimators (i.e., estimators which approximate the solution of Fn([theta]) = 0). It is proved that this class is nonvoid if appropriate regularity conditions are fulfilled and if one has at hand a suitable initial estimator. Furthermore, it is shown that Fn-estimators admit a stochastic expansion (which enables to give results on asymptotic expansions for the distribution of these estimators).
Year of publication: |
1977
|
---|---|
Authors: | Michel, R. |
Published in: |
Journal of Multivariate Analysis. - Elsevier, ISSN 0047-259X. - Vol. 7.1977, 2, p. 235-248
|
Publisher: |
Elsevier |
Keywords: | Iteration procedure stochastic expansions maximum likelihood estimators expansion for the distribution |
Saved in:
Online Resource
Saved in favorites
Similar items by person
-
Michel, R., (1970)
-
Third order efficiency of conditional tests for exponential families
Michel, R., (1979)
-
On Berry-Esseen results for the compound Poisson distribution
Michel, R., (1993)
- More ...