Local influence analysis in the growth curve model with Rao's simple covariance structure
In this paper we discuss the likelihood-based local influence in a growth curve model with Rao's simple covariance structure. Under an abstract perturbation, the Hessian matrix is provided in which the eigenvector corresponding to the maximum absolute eigenvalue is used to assess the influence of observations. Specifically, we employ covariance-weighted perturbation to demonstrate the use of the proposed approach. A practical example is analysed using the proposed local influence approach.
Year of publication: |
2003
|
---|---|
Authors: | Pan, Jian-Xin ; Bai, Peng |
Published in: |
Journal of Applied Statistics. - Taylor & Francis Journals, ISSN 0266-4763. - Vol. 30.2003, 7, p. 771-781
|
Publisher: |
Taylor & Francis Journals |
Saved in:
Saved in favorites
Similar items by person
-
Pan, Jian-Xin, (2000)
-
Multiple outlier detection in growth curve model with unstructured covariance matrix
Pan, Jian-Xin, (1995)
-
Bayesian Influence Assessment in the Growth Curve Model with Unstructured Covariance
Pan, Jian-Xin, (2000)
- More ...