In this paper we survey methods to control for regression model error that is correlated within groups or clusters, but is uncorrelated across groups or clusters. Then failure to control for the clustering can lead to understatement of standard errors and overstatement of statistical...

In this paper we survey methods to control for regression model error that is correlated within groups or clusters, but is uncorrelated across groups or clusters. Then failure to control for the clustering can lead to understatement of standard errors and overstatement of statistical...

We study the impact of graduating in a recession in Flanders (Belgium), i.e. in a rigid labor market. In the presence of a high minimum wage, a typical recession hardly influences the hourly wage of low educated men, but reduces working time and earnings by about 4.5% up to twelve years after...

We study the impact of graduating in a recession in Flanders (Belgium), i.e. in a rigid labor market. In the presence of a high minimum wage, a typical recession hardly influences the hourly wage of low educated men, but reduces working time and earnings by about 4.5% up to twelve years after...

A growing literature on inference in difference-in-differences (DiD) designs with grouped errors has been pessimistic about obtaining hypothesis tests of the correct size, particularly with few groups. We provide Monte Carlo evidence for three points: (i) it is possible to obtain tests of the...

Climate change causes natural disasters to occur at higher frequency and increased severity. Using a unique dataset on German banks, this paper explores how regionally less diversified banks in Germany adjusted their loan loss provisioning following the severe summer flood of 2013, which...

We provide new and computationally attractive methods, based on jackknifing by cluster, to obtain cluster-robust variance matrix estimators (CRVEs) for linear regres- sion models estimated by least squares. These estimators have previously been com- putationally infeasible except for small...

We study cluster-robust inference for binary response models. Inference based on the most commonly-used cluster-robust variance matrix estimator (CRVE) can be very unreliable. We study several alternatives. Conceptually the simplest of these, but also the most computationally demanding, involves...

For linear regression models with cross-section or panel data, it is natural to assume that the disturbances are clustered in two dimensions. However, the finite-sample properties of two-way cluster-robust tests and confidence intervals are often poor. We discuss several ways to improve...

This paper re-examines inference for cluster samples. Sensitivity analysis is proposed as a new method to perform inference when the number of groups is small. Based on estimations using disaggregated data, the sensitivity of the standard errors with respect to the variance of the cluster...