Assessing the nature of periodontal disease progression - an application of covariance structure estimation.
Differing theories of the nature of periodontal disease progression have been proposed, but there is no convincing statistical evidence with which to choose between them.
Three models for disease progression are developed, and their covariance structures are derived and shown to be hierarchical.
The estimation of the covariance structures of the three models is illustrated with both simulated data and data from a long-term study of periodontal disease progression.
The study provides insight into both the nature of periodontal disease progression and the use of the estimation of covariance structural models to distinguish between models for which a likelihood is not completely specified.
Mots-clés Pascal : Estimation statistique, Parodontopathie, Dent pathologie, Evolution, Simulation statistique, Etude longitudinale
Mots-clés Pascal anglais : Statistical estimation, Periodontal disease, Dental disease, Evolution, Statistical simulation, Follow up study
Notice produite par :
Inist-CNRS - Institut de l'Information Scientifique et Technique
Cote : 94-0027466
Code Inist : 002B10C02. Création : 199406.