Document Detail


Just-identified versus overidentified two-level hierarchical linear models with missing data.
MedLine Citation:
PMID:  17501944     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
The development of model-based methods for incomplete data has been a seminal contribution to statistical practice. Under the assumption of ignorable missingness, one estimates the joint distribution of the complete data for thetainTheta from the incomplete or observed data y(obs). Many interesting models involve one-to-one transformations of theta. For example, with y(i) approximately N(mu, Sigma) for i= 1, ... , n and theta= (mu, Sigma), an ordinary least squares (OLS) regression model is a one-to-one transformation of theta. Inferences based on such a transformation are equivalent to inferences based on OLS using data multiply imputed from f(y(mis) | y(obs), theta) for missing y(mis). Thus, identification of theta from y(obs) is equivalent to identification of the regression model. In this article, we consider a model for two-level data with continuous outcomes where the observations within each cluster are dependent. The parameters of the hierarchical linear model (HLM) of interest, however, lie in a subspace of Theta in general. This identification of the joint distribution overidentifies the HLM. We show how to characterize the joint distribution so that its parameters are a one-to-one transformation of the parameters of the HLM. This leads to efficient estimation of the HLM from incomplete data using either the transformation method or the method of multiple imputation. The approach allows outcomes and covariates to be missing at either of the two levels, and the HLM of interest can involve the regression of any subset of variables on a disjoint subset of variables conceived as covariates.
Authors:
Yongyun Shin; Stephen W Raudenbush
Related Documents :
18647294 - Weighted wilcoxon-type smoothly clipped absolute deviation method.
14766384 - Modeling nonlinear effects in longitudinal survival data: implications for the physiolo...
11241574 - Bayesian analyses of longitudinal binary data using markov regression models of unknown...
18179354 - Allowing for correlations between correlations in random-effects meta-analysis of corre...
22376144 - State-of-the art comparability of corrected emission spectra - part ii: field laborator...
22898094 - A market systems analysis of the u.s. sport utility vehicle market considering frontal ...
Publication Detail:
Type:  Journal Article     Date:  2007-05-14
Journal Detail:
Title:  Biometrics     Volume:  63     ISSN:  0006-341X     ISO Abbreviation:  Biometrics     Publication Date:  2007 Dec 
Date Detail:
Created Date:  2007-12-14     Completed Date:  2008-01-22     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  0370625     Medline TA:  Biometrics     Country:  United States    
Other Details:
Languages:  eng     Pagination:  1262-8     Citation Subset:  IM    
Affiliation:
University of Michigan, 439 West Hall, 1085 South University, Ann Arbor, Michigan 48109-1107, USA. choil@umich.edu
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Algorithms*
Artifacts*
Computer Simulation
Data Interpretation, Statistical*
Linear Models*
Models, Biological*
Models, Statistical*
Reproducibility of Results
Sample Size*
Sensitivity and Specificity

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Nonparametric bayesian estimation of positive false discovery rates.
Next Document:  On smoothing trends in population index modeling.