Initial value dependence: comparison of two methods for its study

Access & Terms of Use
metadata only access
Abstract
This article is concerned with the comparison of slope estimator precision in regression analysis and the structural relationships approach (e.g., Humak, 1983, ch. 3; Kendall and Stuart, 1977), as it is relevant for their applications when testing for initial value dependence in biomedical and behavioral contexts of repeated assessments (e.g., Blomqvist, 1977; Wall, 1977). As a basis for the comparison of the two methods, the mean square error is adopted. In the general case, it is argued for an informed (data-dependent) choice between regression analysis and the structural relationships approach. For the apparent majority of biomedical and behavioral studies of the phenomenon of initial value depenence, this comparison suggests that structural relationships is the preferable approach leading to more trustworthy substantive conclusions.
Persistent link to this record
DOI
Additional Link
Author(s)
Raykov, T
Penev, Spiridon
Supervisor(s)
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
1996
Resource Type
Journal Article
Degree Type
UNSW Faculty