🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

What drives Edward Norton to study the #economics of #healthcare, and to partner with hospitals via MVC to quantify the $$$ effects of care improvements? Read more from Michigan Public Health, which just gave him its top research award: sph.umich.edu/stories/2019po…

Their name is Mud

"Requiem for Odds Ratios" in @hsr_hret by Edward Norton, Bryan Dowd, @GarridoMelissa, and Matt Maciejewski.

Price reductions from Oregon’s hospital payment caps are associated with reduced enrollee out-of-pocket spending at the point of care and generated savings for the targeted population. ja.ma/3AKnAf6 Roslyn C Murray, PhD Edward Norton Andrew Ryan

When to apply robust standard errors vs vce cluster? Great discussion at CHOP Data and Donuts 📊 🍩 🙏 Dr. Edward Norton !

Partha Deb, Jeff Wooldridge Jeffrey Wooldridge , Jeff Zabel, and I are pleased that our new difference-in-differences paper is available as an NBER working paper.

Our theoretical contribution: a linear-in parameters regression specification with flexible functional form with group by-time treatment effects, two-way fixed effects, and interaction terms yields consistent estimates of heterogeneous treatment effects under general conditions.

We prove that FLEX treatment effect parameter estimates are asymptotically unbiased estimates of the group-time heterogeneous treatment effects that can be obtained in the repeated cross-section setting by an imputation method. Jeffrey Wooldridge

Furthermore, our FLEX estimates are identical to those from the Borusyak et al. (2024) imputation estimator extended to the case that allows for treatment-effect heterogeneity by group and time (not simply cohort and time). Jeffrey Wooldridge

Our approach is simple and transparent. It is easy to know how identification is achieved, which treatment obs. are compared to which control obs., and how many parameters are estimated. We call it FLEX: it is a flexible linear model estimated by OLS with covariates (X).

Here is the link to the Dropbox where you can download the Stata code and data sets. dropbox.com/scl/fi/upitird… Jeffrey Wooldridge

Nick Hagerty Edward Norton No, we're not restricting individual-level heterogeneity. We're making a conditional parallel trends assumption analogous to previous work -- Callaway-Sant'Anna, Borusyak et al., and so on. We actually allow selection based on group, not cohort, which is a bit more general.

Nick Hagerty Edward Norton Our assumptions are stated in terms of a population. If you put an i subscript on everything, TE_it(g) = Y_it(g) - Y_it(inf) are allowed to depend arbtrarily on i. The "lags only" regression is the repeated CS version of the regression I suggested for the panel case.

Cider donuts and propensity score matching! #FallinAnnArbor Every two weeks our analyst meet to learn new methods and review each others code.

He gave the coup de grace to odds ratios and now he's FLEXing on the diff in diff lit. Edward Norton feeling buff in 2024

We’re always trying to push new methods that are more efficient and robust. Today, @UMichCHOPFellow we took a deep dive on, “Entropy Balancing” with Edward Norton !

The Editorial Boards of leading health econ journals are issuing a Joint Statement of Principles on Editorial Independence. In these troubled times some scholars worry their work might suffer ideological attack. We reaffirm our commitment to unfettered scholarship. Please repost.

The editorial boards of seven leading health econ journals are taking a stand today against political influence and ideological attacks on peer review and academia. No matter what you study we want to see your scholarship and will NEVER collaborate with suppression. Please repost