"Handling Nonignorable Dropout in Longitudinal Clinical Trials" Garrett
Fitzmaurice (School of Public Health, Harvard University)
Wednesday December 11 at noon
Center for Basic Research in the Social Sciences
34 Kirkland Street, Room 22
Lunch will be served.
Abstract: In this talk we give an introduction to two common approaches
to handling nonignorable dropout in longitudinal clinical trials
based on selection and pattern mixture models.
Some of the distinct advantages and disadvantages of selection
and pattern mixture models are reviewed and highlighted.
An example, based on data from a clinical trial of contracepting
women, is used to illustrate some of the main ideas.
There is no paper available, however a related paper by the presenter,
"Methods for Handling Dropouts in Longitudinal Clinical Trials" can be
accessed at: www.courses.fas.harvard.edu/~gov3009/handouts/dropouts.pdf
The Research Workshop in Applied Statistics is a year-long forum for
graduate students, faculty, and visiting scholars to present and discuss
statistical innovations and applications in the social sciences. The
Fitzmaurice talk is the final presentation of the Fall 2002 semester.
The first talk of the Spring 2003 semester, "Enhancing the Validity and
Cross-Cultural Comparability of Survey Research" will be given on January
29 by Gary King (Government, Harvard University). For more information or
to subscribe to the permanent workshop list, email corr(a)fas.harvard.edu.
Trade and Inequality in China: a Bayesian Hierarchical Framework. Tao Li
(Government Department, Harvard University).
Wednesday December 4 at noon
Center for Basic Research in the Social Sciences
34 Kirkland Street, Room 22
Lunch will be served.
Abstract:
A Bayesian hierarchical framework is fitted to panel data on Chinese trade
and income inequality to test whether trade has contributed to the
reduction of rural-urban income inequality in China. The Bayesian
hierarchical framework proves useful for studying the large regional
variability within China. The preliminary results are reported and
compared with alternative econometric models. This presentation will be a
working session, with no paper available.
The Research Workshop in Applied Statistics is a year-long forum for
graduate students, faculty, and visiting scholars to present and discuss
statistical innovations and applications in the social sciences. For more
information or to subscribe to the permanent workshop list, email
corr(a)fas.harvard.edu.