We propose a two-stage MRQAP to analyze dynamic network data, within the framework of an equilibrium-correction (EC) model. Extensive simulation results indicate practical relevance of our method and its improvement over standard OLS. An empirical illustration additionally shows that the EC model yields interpretable parameters, in contrast to an unrestricted dynamic model.

Additional Metadata
Keywords cognitive social structures, consistent accuracy, network centrality, structural autocorrelation, two-stage equilibrium model
JEL Statistical Decision Theory; Operations Research (jel C44), Econometric Modeling: Other (jel C59), Business Administration and Business Economics; Marketing; Accounting (jel M), Marketing (jel M31)
Publisher Erasmus Research Institute of Management
Persistent URL hdl.handle.net/1765/100
Series ERIM Report Series Research in Management
Rights Copyright 2001, D. Dekker, P.H.B.F. Franses, D. Krackhardt, This report in the ERIM Report Series Research in Management is intended as a means to communicate the results of recent research to academic colleagues and other interested parties. All reports are considered as preliminary and subject to possibly major revisions. This applies equally to opinions expressed, theories developed, and data used. Therefore, comments and suggestions are welcome and should be directed to the authors.
Citation
Dekker, D.J, Franses, Ph.H.B.F, & Krackhardt, D. (2001). An Equilibrium-Correction Model for Dynamic Network Data (No. ERS-2001-39-MKT). ERIM Report Series Research in Management. Erasmus Research Institute of Management. Retrieved from http://hdl.handle.net/1765/100