Clustering, Spatial Correlations, and Randomization Inference
AbstractIt is a standard practice in regression analyses to allow for clustering in the error covariance matrix if the explanatory variable of interest varies at a more aggregate level (e.g., the state level) than the units of observation (e.g., individuals). Often, however, the structure of the error covariance matrix is more complex, with correlations not vanishing for units in different clusters. Here, we explore the implications of such correlations for the actual and estimated precision of least squares estimators. Our main theoretical result is that with equal-sized clusters, if the covariate of interest is randomly assigned at the cluster level, only accounting for nonzero covariances at the cluster level, and ignoring correlations between clusters as well as differences in within-cluster correlations, leads to valid confidence intervals. However, in the absence of random assignment of the covariates, ignoring general correlation structures may lead to biases in standard errors. We illustrate our findings using the 5% public-use census data. Based on these results, we recommend that researchers, as a matter of routine, explore the extent of spatial correlations in explanatory variables beyond state-level clustering.
Download InfoIf you experience problems downloading a file, check if you have the proper application to view it first. In case of further problems read the IDEAS help page. Note that these files are not on the IDEAS site. Please be patient as the files may be large.
Bibliographic InfoArticle provided by Taylor & Francis Journals in its journal Journal of the American Statistical Association.
Volume (Year): 107 (2012)
Issue (Month): 498 (June)
Contact details of provider:
Web page: http://www.tandfonline.com/UASA20
Other versions of this item:
- Thomas Barrios & Rebecca Diamond & Guido W. Imbens & Michal Kolesar, 2010. "Clustering, Spatial Correlations and Randomization Inference," NBER Working Papers 15760, National Bureau of Economic Research, Inc.
- C01 - Mathematical and Quantitative Methods - - General - - - Econometrics
- C1 - Mathematical and Quantitative Methods - - Econometric and Statistical Methods and Methodology: General
- C31 - Mathematical and Quantitative Methods - - Multiple or Simultaneous Equation Models; Multiple Variables - - - Cross-Sectional Models; Spatial Models; Treatment Effect Models; Quantile Regressions; Social Interaction Models
You can help add them by filling out this form.
CitEc Project, subscribe to its RSS feed for this item.
- Brewer, Mike & Crossley, Thomas F. & Joyce, Robert, 2013. "Inference with Difference-in-Differences Revisited," IZA Discussion Papers 7742, Institute for the Study of Labor (IZA).
- Hugh Gravelle & Rita Santos & Luigi Siciliani & Rosalind Goudie, 2012. "Hospital Quality Competition Under Fixed Prices," Working Papers 080cherp, Centre for Health Economics, University of York.
- Breinlich, Holger & Ottaviano, Gianmarco & Temple, Jonathan, 2013.
"Regional Growth and Regional Decline,"
CEPR Discussion Papers
9568, C.E.P.R. Discussion Papers.
- Holger Breinlich & Gianmarco I. P. Ottaviano & Jonathan R. W. Temple, 2013. "Regional Growth and Regional Decline," CEP Discussion Papers dp1232, Centre for Economic Performance, LSE.
- Holger Breinlich & Gianmarco I. P. Ottaviano & Jonathan R. W. Temple, 2013. "Regional growth and regional decline," LSE Research Online Documents on Economics 51575, London School of Economics and Political Science, LSE Library.
- Holger Breinlich & Gianmarco I.P. Ottaviano & Jonathan R.W. Temple, 2013. "Regional growth and regional decline," Economics Discussion Papers 729, University of Essex, Department of Economics.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Michael McNulty).
If references are entirely missing, you can add them using this form.