Apologies for cross-posting
*********************************************************
DIPARTIMENTO DI SCIENZE STATISTICHE
Universita' degli Studi di Bologna
Via delle Belle Arti, 41 - 40126 Bologna
*********************************************************
Seminar on "Bayesian model averaging for structural equation models"
Prof. DAVID KAPLAN
Department of Educational Psychology
University of Wisconsin, Madison
Thursday April 28, 2016 at 2:30 p.m.
Aula II, Department of Statistical Sciences
via delle Belle Arti, 41 - Bologna
Abstract:
This talk presents results on the use of Bayesian model averaging as a means of addressing predictive performance in Bayesian structural equation models. The current approach to addressing the problem of model uncertainty from a Bayesian perspective lies in the method of Bayesian model averaging. We expand the work of Madigan and his colleagues by considering a structural equation model as a special case of a directed acyclic graph. We then provide an algorithm that searches the model space for sub-models and obtains a weighted average of the sub-models using posterior model probabilities as weights. Our simulation study provides a frequentist evaluation of our Bayesian model averaging approach and indicates that when the true model is known, Bayesian model averaging does not yield necessarily better predictive performance compared to non-averaged models. However, our case study using data from an international large-scale assessment reveals that the model-averaged sub-models provide better posterior predictive performance compared to the initially specified model.
Anyone interested is invited to participate.
Details: http://www.stat.unibo.it/it/eventi/statistics-seminar-2016-bayesian-model-averaging-for-structural-equation-models
--
Silvia Bianconcini
Dipartimento di Scienze Statistiche
Universita' di Bologna
Via delle Belle Arti 41 - 40126 Bologna, ITALY
tel: +39 051 2098257 fax: +39 051 232153
skype: silviabi2
e-mail: [log in to unmask]
http://www2.stat.unibo.it/bianconcini/
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|