Dr. Joseph Hogan from Brown University presents a lecture titled "Causal Inference"

View Slides

https://drive.google.com/open?id=0B4I...

Lecture Abstract

Discovery of causal relationships is a fundamental objective of scientific research. Statistical methods for causal inference have a long history, but research in the last 30 years in particular has yielded substantial and innovative advances in both theory and methodology. This lecture provides a survey of causal inference from a statistical perspective, with emphasis on methods for observational studies. Part I describes representation of causal effects in terms of potential outcomes and graphical models. Part II uses the simple case of binary treatment to review analytic techniques for inferring causal effects from observed data, including regression adjustment, inverse probability weighting, matching, and standardization. In Part III we discuss more complicated cases where causal questions arise, including time-varying treatment and mediated causal effects. Finally in Part IV we describe some open and active areas of research, such as causal inference on networks, confounder selection in big-data settings, and the role of machine learning in causal inference.

About the Speaker

Joseph Hogan is Professor of Biostatistics and Carole and Lawrence Sirovich Professor of Public Health at Brown University. He also serves as Deputy Director for the Data Science Initiative at Brown. Dr Hogan conducts research on development and application of statistical methods for causal inference and missing data, and has
a long-standing interest in HIV/AIDS. Much of his recent research concerns HIV in Kenya and sub-Saharan Africa. Hogan is co-author (with Michael Daniels) of the book Missing Data in Longitudinal Studies (Chapman & Hall, 2008), and served on the National Research Council panel that generated the report Prevention and Treatment of Missing Data in Clinical Trials (National Academies Press, 2010).

Join our weekly meetings from your computer, tablet or smartphone.

Visit our website to view our schedule and join our next live webinar! http://www.bigdatau.org/data-science-...