|
|||||
|
|
Windows of Opportunity: Window Subseries Empirical Variance Estimators in International Relations |
|
| Authors: | Heagerty, Patrick Ward, Michael D. Gleditsch, Kristian Skrede |
| Affiliation: | Department of Biostatistics, University of Washington, Seattle, WA 98195 e-mail: heagerty{at}u.washington.edu Department of Political Science and Center for Statistics in the Social Sciences, University of Washington, Seattle, WA 98195, and Éspace Éurope, Université Pierre Mendès France, Grenoble, France, BP 38040 e-mail: mdw{at}u.washington.edu Department of Political Science, University of California, San Diego, La Jolla, CA 92093-0521 e-mail: kgleditsch{at}ucsd.edu |
| Abstract: | We show that temporal, spatial, and dyadic dependencies amongobservations complicate the estimation of covariance structuresin panel databases. Ignoring these dependencies results in covarianceestimates that are often too small and inferences that may bemore confident about empirical patterns than is justified bythe data. In this article, we detail the development of a nonparametricapproach, window subseries empirical variance estimators (WSEV),that can more fully capture the impact of these dependencieson the covariance structure. We illustrate this approach ina simulation as well as with a statistical model of internationalconflict similar to many applications in the international relationsliterature. |
| Keywords: | |
| 本文献已被 Oxford 等数据库收录！ | |