« Back to Results

Inference with Dependent Data

Paper Session

Sunday, Jan. 8, 2023 1:00 PM - 3:00 PM (CST)

Hilton Riverside, Trafalgar
Hosted By: Econometric Society
  • Chair: Christian B. Hansen, University of Chicago

Standard Errors for Two-Way Clustering with Serially Correlated Time Effects

Harold D. Chiang
,
University of Wisconsin-Madison
Bruce Hansen
,
University of Wisconsin
Yuya Sasaki
,
Vanderbilt University

Abstract

We propose improved standard errors and an asymptotic distribution theory for two-way clustered panels. Our proposed estimator and theory allow for arbitrary serial dependence in the common time effects, which is excluded by existing two-way methods, including the popular two-way cluster standard errors of Cameron, Gelbach, and Miller (2011) and the cluster bootstrap of Menzel (2021). Our asymptotic distribution theory is the first which allows for this level of interdependence among the observations. Under weak regularity conditions, we demonstrate that the least squares estimator is asymptotically normal, our proposed variance estimator is consistent, and t-ratios are asymptotically standard normal, permitting conventional inference. We present simulation evidence that confidence intervals constructed with our proposed standard errors obtain superior coverage performance relative to existing methods. We illustrate the relevance of the proposed method in an empirical application to a standard Fama-French three-factor regression.

When Do State-Dependent Local Projections Work?

Silvia Goncalves
,
McGill University
Ana Maria Herrera
,
University of Kentucky
Lutz Kilian
,
Federal Reserve Bank of Dallas
Elena Pesavento
,
Emory University

Abstract

Many empirical studies estimate impulse response functions that depend on the state of the economy. Most of these studies rely on a variant of the local projection (LP) approach to estimate the state-dependent impulse response functions. Despite its widespread application, the asymptotic validity of the LP approach to estimating state-dependent impulse responses has not been established to date. We formally derive this result for a structural state-dependent vector autoregressive process. The model only requires the structural shock of interest to be identified. A crucial condition for the consistency of the state-dependent LP estimator of the response function is that current and future states are conditionally mean independent of the structural shocks, given the information available at the time the shock is realized. This rules out models in which the state of the economy is a function of current or future realizations of the outcome variable of interest, as is often the case in applied work. Even when the state is a function of past values of this variable only, consistency may hold only at short horizons.

A Design-Based Approach to Spatial Correlation

Ruonan Xu
,
Rutgers University
Jeffrey M. Wooldridge
,
Michigan State University

Abstract

When observing spatial data, what standard errors should we report? With the finite population framework, we identify three channels of spatial correlation: sampling scheme, assignment design, and model specification. The Eicker-Huber-White standard error, the cluster-robust standard error, and the spatial heteroskedasticity and autocorrelation consistent (SHAC) standard error are compared under different combinations of the three channels. Then, we provide guidelines for whether standard errors should be adjusted for spatial correlation for both linear and nonlinear estimators. As it turns out, the answer to this question also depends on the magnitude of the sampling probability.

Wild Bootstrap for Instrumental Variables Regressions with Weak and Few Clusters

Wenjie Wang
,
Nanyang Technological University
Yichong Zhang
,
Singapore Management University

Abstract

We study the wild bootstrap inference for instrumental variable regressions in the framework of a small number of large clusters in which the number of clusters is viewed as fixed and the number of observations for each cluster diverges to infinity. We first show that the wild bootstrap Wald test, with or without using the cluster-robust covariance estimator, controls size asymptotically up to a small error as long as the parameters of endogenous variables are strongly identified in at least one of the clusters. Then, we establish the required number of strong clusters for the test to have power against local alternatives. We further develop a wild bootstrap Anderson-Rubin test for the full-vector inference and show that it controls size asymptotically up to a small error even under weak or partial identification in all clusters. We illustrate the good finite sample performance of the new inference methods using simulations and provide an empirical application to a well-known dataset about US local labor markets.

Inference for Cluster Randomized Experiments with Non-ignorable Cluster Sizes

Federico Andres Bugni
,
Northwestern University
Ivan Alexis Canay
,
Northwestern University
Azeem M. Shaikh
,
University of Chicago
Max Tabord-Meehan
,
University of Chicago

Abstract

This paper considers the problem of inference in cluster randomized experiments when cluster sizes are non-ignorable. Here, by a cluster randomized experiment, we mean one in which treatment is assigned at the level of the cluster; by non-ignorable cluster sizes we mean that "large" clusters and "small" clusters may be heterogeneous, and, in particular, the effects of the treatment may vary across clusters of differing sizes. In order to permit this sort of flexibility, we consider a sampling framework in which cluster sizes themselves are random. In this way, our analysis departs from earlier analyses of cluster randomized experiments in which cluster sizes are treated as non-random. We distinguish between two different parameters of interest: the equally-weighted cluster-level average treatment effect, and the size- weighted cluster-level average treatment effect. For each parameter, we provide methods for inference in an asymptotic framework where the number of clusters tends to infinity and treatment is assigned using simple random sampling. We additionally permit the experimenter to sample only a subset of the units within each cluster rather than the entire cluster and demonstrate the implications of such sampling for some commonly used estimators. A small simulation study shows the practical relevance of our theoretical results.
JEL Classifications
  • C1 - Econometric and Statistical Methods and Methodology: General
  • C3 - Multiple or Simultaneous Equation Models; Multiple Variables