PPT-1 Big Data, Causal Inference, and Instrumental Variables

Center for Health Policy and Healthcare Research January 22 2015 Steven D Pizer PhD Associate Professor of Health Economics Department of Pharmacy and Health Systems Sciences School of . Technical Track Session IV. This . material constitutes supporting material for the "Impact Evaluation in Practice" book. This additional material is made freely but please acknowledge its use as follows: . As some see it…. • “Instrumental music is much more moving.”. • “Many enjoy it more than singing alone.” . • “It is a mark of growing progressive modern churches.”. • “The Bible doesn’t say you can’t use instruments.”. Outline 1. Comparison of Classical and Instrumental Conditioning. Early Investigations of Instrumental Conditioning.. Thorndike. Chicks and mazes. Cats and puzzle box. Modern approaches to the study of instrumental conditioning. Michael Rosenblum. March 16, 2010. Overview. I describe the set of assumptions encoded by a causal directed acyclic graph (DAG). I use an example from page 15 of the book . Causality. by Judea Pearl (2009). . Susan Athey, Stanford GSB. Based on joint work with Guido Imbens, Stefan Wager. References outside CS literature. Imbens and Rubin Causal Inference book (2015): synthesis of literature prior to big data/ML. Naftali Weinberger. Tilburg Center for Logic, Ethics and Philosophy of Science. Time and Causality in the Sciences. June 8. th. , 2017. Principle of the . C. ommon Cause. iPad. Happiness. iPad. Happiness. . Richard Scheines. Philosophy, Machine Learning, . Human-Computer Interaction . Carnegie Mellon University. 2. Goals. Basic Familiarity with Causal Model Search: . What it is. What it can and cannot do. Tony Cox. May 5, 2016. 1. Download free CAT software from: . http://cox-associates.com/CAT.htm. . Outline. Why CAT? Challenges for causal analytics. Ambiguous C-R associations: theory & practice. Austin Nichols (Abt) & Linden McBride (Cornell). July 27, 2017. Stata Conference. Baltimore, MD. Overview. Machine learning methods dominant for classification/prediction problems.. Prediction is useful for causal inference if one is trying to predict propensity scores (probability of treatment conditional on observables);. Sciences: QUICK EXAMPLES. #. konfoundit. Kenneth A. . Frank. Ran . Xu; Zixi . Chen. ; I-Chien Chen, Guan Saw. 2018. (. AERA on-line video – cost is . $105. ). Motivation . Statistical inferences are often challenged because of uncontrolled bias. There may be bias due to uncontrolled confounding . Alex . Dimakis. UT Austin. j. oint work with . Murat . Kocaoglu. , . Karthik. . Shanmugam. Sriram. . Vishwanath. , . Babak. . Hassibi. Overview. Discovering causal directions . Part 1: . Interventions. Instrumental variables IVs are used to control for confounding and measurement error in ssibility of making causal inferences with observational data Like propensity scores IVconfounding effects Other Obid. . A.Khakimov. Revew. . Four complications that induce correlation between . X. and . e. Omitted Variables Bias. Measurement Error. Simultaneous Causality. Using Lagged Values of the Dependent Variable as . (. CCD. ). of Biomedical Knowledge from Big . Data. University of Pittsburgh. Carnegie Mellon . University. Pittsburgh Supercomputing . Center. Yale . University. PIs: . Greg Cooper, Ivet . Bahar, Jeremy Berg.