What to Do (and Not Do) with Time-Series Criss-Section Data
American Political Science Review, Vol. 89, No. 3, September 1995
14 Pages Posted: 15 Aug 2010
Date Written: August 13, 2010
Abstract
We examine some issues in the estimation of time-series cross-section models, calling into question the conclusions of many published studies, particularly in the field of comparative political economy. We show that the generalized least squares approach of Parks produces standard errors that lead to extreme overconfidence, often underestimating variability by 50% or more. We also provide an alternative estimator of the standard errors that is correct when the error structures show complications found in this type of model. Monte Carlo analysis shows that these “panel-corrected standard errors” perform well. The utility of our approach is demonstrated via a reanalysis of one “social democratic corporatist” model.
Keywords: panel, time-series cross-section, econometrics
Suggested Citation: Suggested Citation
Do you have a job opening that you would like to promote on SSRN?
Recommended Papers
-
Bailing Out Expectations and Health Expenditure in Italy
By Massimo Bordignon and Gilberto Turati
-
Social Democrats and the New Partisan Politics of Public Investment in Education
-
Fixed Effects Vector Decomposition: Reply
By Thomas Pluemper and Vera E. Troeger
-
Opinion Polarization and Inter-Party Competition on Europe
By Carole J. Wilson and Ian Down
-
Bootstrap Methods for Inference in a Sur Model with Autocorrelated Disturbances
-
Strange Bedfellows: The Policy Consequences of Legislative-Judicial Relations in the American States