image image
Media

Document Detail


permalink to this item
WORKING PAPER
Spatio-Temporal Models for Political-Science Panel and Time-Series-Cross-Section Data
Franzese, Robert
Hays, Jude

Abstract
Building from our broader project exploring spatial-econometric models for political science, this paper discusses estimation, interpretation, and presentation of spatio-temporal models. We first present a generic spatio-temporal-lag model and two methods, OLS and ML, for estimating parameters in such models. We briefly consider those estimators’ properties analytically before showing next how to calculate and to present the spatio-temporal dynamic and long-run steady-state equilibrium effects—i.e., the spatio-temporal substance of the model—implied by the coefficient estimates. Then, we conduct Monte Carlo experiments to explore the properties of the OLS and ML estimators, and, finally, we conclude with a reanalysis of Beck, Gleditsch, and Beardsley’s (2006) state-of-the-art study of directed export flows among major powers.

Keywords
Panel Data
Spatial Econometrics
Spatial-Lag Model
Spatio-Temporal Dynamics
Spatio-Temporal Model
Spatio-Temporal Multiplier
Spatio-Temporal Steady-State Effects
Time-Series-Cross-Section Data


File
icnPdfMini FranzeseHays.PolMeth06.pdf


Uploaded
07-18-2006

Document ID Number
626


   
wustlArtSci