Works in Progress

Judge Influence and Judicial Network

I study the extent to which judges’ influence depend on the judges’ social connections. Guided by a theoretical model that formalizes the role of social connection, I document that social connections are a significant determinant of judge influence. I use the flow of law clerks between judges from 1995-2004 as a measure of social connections, total citations as a proxy for influence, and I address network endogeneity by using novel data on the judges’ alumni connections. Our results also provide new insights into how social connectedness interacts with judge demographic characteristics.

Policy Diffusion Networks and Campaign Donations: Evidence from Text Reuse in State Legislatures

Dynamic network formation with forward looking agents with Marco Battaglini and Eleonora Patacchini


Finite-sample inference and nonstandard asymptotics with Monte Carlo tests and R
with Jean-Marie Dufour Handbook of Statistics Volume 41, 2019, Pages 3-31

We review the concept of Monte Carlo test as a simulation-based inference procedure which allows one to construct tests with provably exact levels in situations where the distribution of a test statistic is difficult to establish but can be simulated. The number of simulations required can be extremely small, as low as 19 to run a test with level 0.05. We discuss three extensions of the method: (1) a randomized tie-breaking technique which allows one to use test statistics with discrete null distributions, without further information on the mass points; (2) an extension (maximized Monte Carlo tests) which yields provably valid tests when the test statistic depends on a (finite) number of nuisance parameters; (3) an asymptotic version which allows one to get asymptotically valid tests without any need to establish an asymptotic distribution. As the method is computer intensive, we describe an R package (MaxMC) that allows one to implement this type of procedure. A number of special cases and applications are discussed.