This paper proposes a simple method for testing whether non-compliance in experiments is ignorable, i.e., not jointly related to the treatment and the outcome. The approach consists of (i) regressing the outcome variable on a constant, the treatment, the assignment indicator, and the...

The nonparametric identification of the local average treatment effect (LATE) hinges on the satisfaction of three instrumental variable assumptions: (1) unconfounded assignment of the instrument, (2) no average direct effect of the instrument on the outcome within compliance types (exclusion...

In this paper, we derive a class of modified score tests robust to local and distributional misspecifications for testing spatial error autocorrelation and spatial lag dependence. The proposed tests are general enough to include several popular tests for the spatial dependence as special cases....

In this paper, several tests are suggested for the existence of individual and time effects in panel data models with interactive effects. Their asymptotic properties are obtained under some mild conditions. Monte Carlo simulation is carried out for illustration.

We extend Hansen’s (2005) test to testing hypotheses involving general inequality constraints where the variance–covariance matrix of the functions in the constraints depends on the unknown parameters. The test can be applied to a wider class of problems than Wolak’s (1991).

A simple example is provided that shows that one test can strictly dominate another test in both local power and its robustness to asymptotic size distortion under local model misspecification, with both tests having asymptotic size equal to nominal size under correct model specification....

This note axiomatizes the incomplete preference ordering that reflects statewise dominance with respect to expected utility, as well as the according choice correspondence. The main motivation is to clarify how admissibility as understood by statisticians relates to existing axiomatizations. The...

We show that confidence regions covering the identified set may be preferable to confidence regions covering each of its points in robust control applications.

We derive the limiting distribution of the Oaxaca estimator of average treatment effects studied by Kline (2011). A consistent estimator of the asymptotic variance is proposed that makes use of standard regression routines. It is shown that ignoring uncertainty in group means will tend to lead...

This paper compares asymptotic and finite sample properties of linear IV and bivariate probit in models with an endogenous binary treatment and binary outcome. The results provide guidance on the choice of model specification and help to explain large differences in the estimates depending on...