Do dynamics and heterogeneity in panel data models matter?

Autor(en): Steiner, Andreas
Stichwörter: BIAS; Business & Economics; Central Bank Behaviour; DEMAND; Economics; ESTIMATORS; INTERNATIONAL RESERVES; Mathematical Methods In Social Sciences; Panel data models; Parameter heterogeneity; Poolability; RESTRICTIONS; Social Sciences, Mathematical Methods; TESTS
Erscheinungsdatum: 2011
Herausgeber: PHYSICA-VERLAG GMBH & CO
Journal: EMPIRICAL ECONOMICS
Volumen: 40
Ausgabe: 1
Startseite: 165
Seitenende: 176
Zusammenfassung: 
This article investigates empirically the determinants of central banks' reserve holdings for a large panel data set of developing and transition countries covering the period from 1975 to 2003. It estimates both a static and a dynamic relationship and applies estimators for homogeneous and heterogeneous panel data. Thereby, it examines the extent to which conclusions of panel data studies on the determinants of international reserve holdings are robust to the inclusion of dynamics as well as to the consideration of heterogeneity across countries. The results show that the neglect of dynamics and heterogeneity in country behaviour may lead to misleading inferences. Independently of the chosen estimation method, the findings suggest that trade openness and external debt are robust determinants of the level of reserves. Central banks take precautionary measures against the downside of the increasing international economic integration.
ISSN: 03777332
DOI: 10.1007/s00181-010-0423-2

Show full item record

Google ScholarTM

Check

Altmetric