Analysis of Rank Measures of Association for Ordinal Data from Longitudinal Studies

Autor:	Kerry B. Hafner, Gary G. Koch, Gregory J. Carr
Rok vydání:	1989
Předmět:	Statistics and Probability Ordinal data Rank (linear algebra) Group (mathematics) Covariance matrix Statistics Econometrics Repeated measures design Statistics Probability and Uncertainty Missing data Ordinal response Variable (mathematics) Mathematics
Zdroj:	Journal of the American Statistical Association. 84:797-804
ISSN:	1537-274X 0162-1459
DOI:	10.1080/01621459.1989.10478840
Popis:	Consideration is given to longitudinal data settings in which an ordinal response variable is observed at two or more visits for each subject in two or more ordered groups with at least moderately large sample sizes (e.g., overall n ≥ 40). Rank measures of association between group and response are constructed at each time, and the covariance matrix of these measures is also estimated. Variation among the rank measures of association is then analyzed through weighted least squares methods with weights based on the estimated covariance matrix. Such methods permit evaluation of group × visit interaction as well as other effects of interest. Extensions of this type of analysis to situations with stratification, or concomitant variables, or missing data are also considered. Examples are provided for illustrative purposes.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::60b3557121d16349552f96b1ea662387 https://doi.org/10.1080/01621459.1989.10478840 Zobrazit plný text záznamu