|
 |
|
|
Lin and Colleagues Propose Latent Pattern Mixture Model in Biometrics
Lead Article
 In the lead article in the June issue of Biometrics, Haiqun Lin,
Assistant Professor of Public Health in the Division of Biostatistics,
first author, and co-authors Charles McCulloch of the Division of Biostatistics
at the University of California, San Francisco and Robert Rosenheck, Professor
of Psychiatry and Public Health at Yale, propose a latent pattern mixture
model designed to allow researchers to handle arbitrary patterns of missing
data caused by both subjects’ failure to appear for scheduled visits
and their appearance for unscheduled visits. Although the statistical
literature has not extensively discussed the handling of such intermittent
missing data, it is important to take it into consideration because failure
to do so may cause serious bias in evaluating the effectiveness of randomized
treatments. The article shows how missing data affected the evaluation
of mental health and housing outcomes in three different housing interventions
conducted by the U.S. Department of Veterans Affairs for homeless veterans
with mental illness.
For more information, please link
to the article.
|
