Biostatistics Seminar guest speaker, Michael Wallace, University of Waterloo, will present, "Dynamic treatment regimes and reward ignorant modelling".
Personalized medicine optimizes patient outcome by tailoring treatments to patient-level characteristics. This approach is formalized by dynamic treatment regimes (DTRs): decision rules that take patient information as input and output recommended treatment decisions. The DTR literature has seen the development of increasingly sophisticated causal inference techniques, which attempt to address the limitations of our typically observational datsaets. We note, however, that in practice most patients should often receive optimal or near-optimal treatment, and so the outcome used as part of a typical DTR analysis does not provide much additional information. In light of this, we propose reward ignorant modelling: ignoring the outcome and eliciting an optimal DTR by regressing the observed treatment on relevant covariates, as in a more standard analysis. We present some results from investigating this concept, and analysis of data from the International Warfarin Pharmacogenetics Consortium.
Last Updated On Friday, October 27, 2017 by Haydo, Amber LC
Created On Wednesday, August 09, 2017
Zeda Li, Paul H. Chook Department of Information Systems and Statistics, City University of New York, will present, “Adaptive Bayesian Time-Frequency Analysis of Multivariate Time Series”.
For information on seminars and events in the department, contact:
Update Your Profile