On estimation and testing when explanatory variables are partly endogenous

https://doi.org/10.1016/0304-4076(94)90093-0Get rights and content

Abstract

When some of explanatory variables are correlated with regression errors, the standard econometric theory advocates the use of instrumental variables estimators. This paper considers a special case where applied researchers do not have a sufficient number of instruments but instead they do have prior information on parameters in the form of parameter restrictions. The paper proposes a simple method of estimation and hypothesis testing in the above context. The results are extended to restricted GMM estimation in nonlinear context.

References (22)

  • C.L.F. Attfield

    Estimation and testing when explanatory variables are endogenous

    Journal of Econometrics

    (1991)
  • R.E. Cumby et al.

    Two-step two-stage least squares estimation in models with rational expectations

    Journal of Econometrics

    (1983)
  • J. Dreze et al.

    Bayesian analysis of simultaneous equations systems

  • J. Aitchison et al.

    Maximum-likelihood estimation procedures and associated tests of significance

    Journal of the Royal Statistical Society B

    (1960)
  • T. Amemiya

    Advanced econometrics

    (1985)
  • D.W.K. Andrews

    Asymptotic results for generalized Wald tests

    Econometric Theory

    (1987)
  • D.W.K. Andrews

    Consistency in nonlinear econometric models: A generic uniform law of large numbers

    Econometrica

    (1987)
  • D.W.K. Andrews

    Heteroskedasticity and autocorrelation consistent covariance matrix estimation

    Econometrica

    (1991)
  • T. Breusch

    Hypothesis testing in unidentified models

    Review of Economic Studies

    (1986)
  • R.S. Cantrell et al.

    Interpretation and use of generalized Chow tests

    International Economic Review

    (1991)
  • A.R. Gallant

    Nonlinear statistical models

    (1987)
  • Cited by (0)

    The author would like to thank De-Min Wu, an associate editor, and two anonymous referees for their helpful comments and suggestions.

    View full text