TY - JOUR
T1 - Heuristic Optimization Methods for Dynamic Panel Data Model Selection
T2 - Application on the Russian Innovative Performance
AU - Savin, Ivan
AU - Winker, Peter
N1 - Funding Information:
Acknowledgments Valuable comments and suggestions from Christian Borgelt and Frank Windmeijer are gratefully acknowledged. All remaining shortcomings are our responsibility. Financial support from the German Academic Exchange Service (DAAD) and the EU Commission through MRTN-CT-2006-034270 COMISEF is gratefully acknowledged.
PY - 2012/4
Y1 - 2012/4
N2 - Innovations, be they radical new products or technology improvements, are widely recognized as a key factor of economic growth. To identify the factors triggering innovative activities is a main concern for economic theory and empirical analysis. As the number of hypotheses is large, the process of model selection becomes a crucial part of the empirical implementation. The problem is complicated by unobserved heterogeneity and possible endogeneity of regressors. A new efficient solution to this problem is suggested, applying optimization heuristics, which exploits the inherent discrete nature of the model selection problem. The method is applied to Russian regional data within the framework of a log-linear dynamic panel data model. To illustrate the performance of the method, we also report the results of Monte-Carlo simulations.
AB - Innovations, be they radical new products or technology improvements, are widely recognized as a key factor of economic growth. To identify the factors triggering innovative activities is a main concern for economic theory and empirical analysis. As the number of hypotheses is large, the process of model selection becomes a crucial part of the empirical implementation. The problem is complicated by unobserved heterogeneity and possible endogeneity of regressors. A new efficient solution to this problem is suggested, applying optimization heuristics, which exploits the inherent discrete nature of the model selection problem. The method is applied to Russian regional data within the framework of a log-linear dynamic panel data model. To illustrate the performance of the method, we also report the results of Monte-Carlo simulations.
KW - Dynamic panel data
KW - Genetic algorithms
KW - GMM
KW - Innovation
KW - Model selection
KW - Threshold accepting
UR - https://www.scopus.com/pages/publications/84858291916
U2 - 10.1007/s10614-010-9243-x
DO - 10.1007/s10614-010-9243-x
M3 - Article
AN - SCOPUS:84858291916
SN - 0927-7099
VL - 39
SP - 337
EP - 363
JO - Computational Economics
JF - Computational Economics
IS - 4
ER -
