The Asymptotic Unbiasedness of $S^2$ in the Linear Regression Model with Moving Average or Particular S-th Order Autocorrelated Disturbances

  • Song, Seuck-Heun (Department of Statistics, Dortmund University, Dortmund 44221, Germany)
  • Published : 1994.06.01

Abstract

The OLS-estimator of the distribance variance in the linear regression model is shown to be asymptotically unbiased when the disturbances are MA(1)-process or particular s-th autocorrelated AR(s)-process.

Keywords

References

  1. The American Statistician v.40 Bias of s² in Linear Regression with Dependent Errors Dufour,J.M.
  2. Matrices with Applications in Statistics Graybill,F.A.
  3. Topics in Matrix Analysis Horn,R.A.;Johnson,C.R.
  4. Journal of Econometrics v.24 A New Testn for Fourth-Order Autoregressive Disturvances King,M.J.
  5. Statistical Papers v.32 The asymototic unbiasedness of s² in the Linear Regression Model with AR(1)-disturbances Kramer,W.
  6. Econometrica v.45 Bounds for the Bias of the Least Squares Estimator of σ² in case of a First-Order Autoregressive Process(Positive Autocorrelation) Neudecker,H.
  7. Econometrica v.46 Bounds for the Bias of the LS Estimator in case of a First-Order(positive) Autoregressive Process Where Regression Contains a Constant Term Neudecker,H.
  8. Econometrica v.42 Bounds on the Variance of Regression Coefficients due to Heteroscedastic or Autoregressive Errors Sathe,S.T.;Vinod,H.D.
  9. Journal of the Royal Statistical Society A v.134 Seasonal Variation in Regression Analysis Thomas,J.J.;Wallis,K.F.
  10. Econometrica v.40 Testing for Fourth Order Autocorrelation in quarterly Regression Equations Wallis,K.F.