About this site
This is the most up-to-date version of my academic website. I'm an associate professor at the University of Colorado Anschutz Medical Campus (more about me on the About page). I have another website that you can access here, which is hosted by the economics department, where I have a courtesy appointment, but I'm not updating that site anymore.
See the Book page for updates on my forthcoming book on master's/PhD-level methods for health services research and health economics to be published by Cambridge University Press. Some sample chapters are available on this site and by request. The first part of the book deals with parametric and nonparametric model estimation and inference (aka, hypothesis testing), including linear regression/OLS, maximum likelihood estimation, GLMs, and marginal effects to interpret model results. The first part also introduces the potential outcomes framework to conceptualize causality as a prediction problem dealing with missing data. The second part covers the most common methods when ignorability (or the conditional independence assumption) does not hold: difference-in-differences, regression discontinuity, instrumental variables, and a sprinkle of longitudinal data (it helps, but not much). Read the Preface to understand more about the book approach and teaching philosophy. Everything in the book can be replicated in Stata and R. I'm adding some Python as well for an appendix on machine learning for prediction. Although causality is about predicting unobserved counterfactuals, machine learning is still not a tool used in causal inference (largely because we need hypothesis testing in explanatory academic research). Still, there are some interesting approaches incorporating machine learning tools in causal inference and tools for prediction in the context of causal inference.
Over the years, I have coded several software tools. I don't update them anymore but they are still being used (based on questions I get). They are in the Code page. It's mostly Stata and SAS. I'll be adding more Python in the coming months as I work on a machine learning appendix.
Email is the best way to get in touch with me. Although I can't answer all emails (but I try), I appreciate the thank-you notes and questions. I'm glad the material is helpful. Do keep in mind the copyright notice, though. I've had bad experiences in the past. It's not pleasant to find your work on somebody else's book.