Personalised Treatment: Learning and Decision

Empirical evidence in social and biomedical sciences commonly suggests individual's response to public policy or medical treatment is heterogeneous. How to efficiently learn and exploit such heterogeneity for the purpose of designing personalised policy/treatment are important topics of interdisciplinary interests.

This one-day workshop presents recent developments on evidence-based design of personalized treatment and targeting policies. It aims to offer the researchers in economics and other disciplines (statistics, medicine, epidemiology, etc.) an opportunity to discuss the issues and ideas on some of the common topics, including:

• Econometric and machine learning methods for personalised treatment/policy
• Medical or policy decision under ambiguity
• Meta-analysis for medical or policy decision making

Programme

Cemmap/UCL Workshop

Personalised Treatment: Learning and Decision

26^nd March (Tue) 2019 at CILIP conference room, 1^st floor, 7 Ridgmount Street.

9:50 – 10:20	Coffee and registration
Session 1:
10:20 – 11:10	Rachel Cassidy (IFS) “Tuberculosis Diagnosis and Treatment under Ambiguity” (with C. Manski)
11:10 – 12:00	Charles Manski (Northwestern) “Meta-Analysis for Medical Decisions”
12:00 – 14:00	Lunch and Coffee
(12:30 – 13:30)	Cemmap seminar by Shosei Sakaguchi (UCL), IFS basement seminar room “Estimating Optimal Dynamic Treatment Assignment Rules under Intertemporal Budget Constraint”
Session 2:
14:00 – 14:50	Stefan Wager (Stanford GSB) “A Robust Method for Dynamic Treatment Choice” (with X. Nie and E. Brunskill)
14:50 – 15:40	Jason Abaluck (Yale SoM) “Who Should Get Blood? Personalizing Medicine with HeterogeneousTreatment Effects”
15:40 – 16:10	Coffee
Session 3:
16:10 – 17:00	Sukjin Han (Texas Austin) “Identification in Nonparametric Models for Dynamic Treatment Effects”
17:00 – 17:50	Karun Adusumilli (UPenn) “Dynamically Optimal Treatment Allocation using Reinforcement Learning” (with F. Geiecke and C. Schilter)