Registration for the Best Methods Bootcamp is now open, and space is limited.
In five days, Brent Goldfarb, Dave Waguespack, and I will help you build a clear, practical toolkit for doing and evaluating empirical research-especially work making causal claims. By the end of the bootcamp, you'll be able to diagnose why estimates are biased, choose appropriate identification strategies, and clearly articulate when and why each method works.
When: June 1–5, 9:30 AM – 2:30 PM EDT (via Zoom)
Apply by: May 21, 2026 (application form here; registration link here)
The bootcamp is designed for PhD students in management and strategy, but all are welcome. It is especially useful for early-stage PhDs who want to "get" empirical methods rather than just run them.
Our approach is problem-based and intuitive. We start with the core question: why do estimates go wrong? You'll learn to recognize and diagnose issues like omitted variables, simultaneity, measurement error, bad controls, and sample selection-and to distinguish between the data generating process, the estimand, and the estimator.
We then walk through the main solutions-instrumental variables, difference-in-differences, discontinuity designs, matching/weighting/doubly robust methods, and double machine learning-with a focus on when they work, when they don't, and how to use them correctly. We also introduce DAGs and simulations to make these ideas more transparent and easier to apply, even if your prior training is limited.
Along the way, we cover practical topics like clustering standard errors, interpreting non-linear models, visualization, and heterogeneous treatment effects.
A syllabus is available here.
If you're looking to build confidence in your empirical work early in your PhD, we'd be glad to have you join us.
Questions can be sent to rhsmith-bestmethods@umd.edu.
Best,
Evan, Brent, and Dave
--