Health Economics Letter

Cluster effects and simultaneity in multilevel models

  1. Richard Blundell1,
  2. Frank Windmeijer2,*

Article first published online: 4 DEC 1998

DOI: 10.1002/(SICI)1099-1050(199707)6:4<439::AID-HEC276>3.0.CO;2-B

Issue

Health Economics

Health Economics

Volume 6, Issue 4, pages 439–443, July 1997

Additional Information

How to Cite

Blundell, R. and Windmeijer, F. (1997), Cluster effects and simultaneity in multilevel models. Health Econ., 6: 439–443. doi: 10.1002/(SICI)1099-1050(199707)6:4<439::AID-HEC276>3.0.CO;2-B

Author Information

  1. 1

    Department of Economics, University College London and Institute for Fiscal Studies, London, UK

  2. 2

    Institute for Fiscal Studies, London, UK

*Institute for Fiscal Studies, 7 Ridgmount Street, London WC1E 7AE, UK.

Publication History

  1. Issue published online: 4 DEC 1998
  2. Article first published online: 4 DEC 1998
  3. Manuscript Accepted: 21 MAR 1997
  4. Manuscript Received: 20 DEC 1996

Funded by

  • King's Fund
  • ESRC Research Centre, Institute for Fiscal Studies

SEARCH

SEARCH BY CITATION

ARTICLE TOOLS

Get PDF (100K)

Keywords:

  • multilevel model;
  • correlated group effects;
  • group size;
  • simultaneity

Abstract

For small group sizes, the GLS estimator in multilevel models is biased and inconsistent when the random cluster effects are correlated with the regressors. A fixed effects approach, conditioning on the cluster effects, provides consistent estimates for the slope parameters. The two estimators are equivalent when group sizes are large. The same results obtain for two-stage estimation procedures that allow for some of the regressors to be simultaneously determined with the dependent variable. The GLS and fixed effects estimators are applied to data on acute care hospital utilization in the UK, allowing for health authority district effects. © 1997 John Wiley & Sons, Ltd.

Get PDF (100K)