A Note on the Asymptotic Property of S2 in Linear Regression Model with Correlated Errors

DOI QR코드

DOI QR Code

Lee, Seung-Chun

  • 발행 : 2003.04.01

초록

An asymptotic property of the ordinary least squares estimator of the disturbance variance is considered in the regression model with correlated errors. It is shown that the convergence in probability of S$^2$ is equivalent to the asymptotic unbiasedness. Beyond the assumption on the design matrix or the variance-covariance matrix of disturbances error, the result is quite general and simplify the earlier results.

키워드

convergence in probability;convergence in L$_1$;uniform integrability asymptotic unbiasedness;strictly stationary process

참고문헌

  1. Empirical Economics v.16 pp.375-377 Consistency of S² in the linear regression model with correlated errors Kramer, W.;Berghoff, S. https://doi.org/10.1007/BF01206283
  2. Introduction to statistical time series Fuller, W. A.
  3. Statistical Papers v.32 pp.71-72 The asymptotic unbiasedness of S² in the linear regression model with AR(1)-disturbances Kramer, W. https://doi.org/10.1007/BF02925481
  4. Journal of the Korean Statistical Society v.25 pp.235-241 The asymptotic unbiasedness of S? in the linear regression model with dependent errors Lee, S.;Kim, Y.-W.
  5. Econometrica v.45 pp.1258-1262 Bounds for the bias of the least squars estimator of σ² in case of a first-order autoregressive process (positive autocorrelation) Neudecker, H.
  6. Journal of the Korean Statistical Society v.23 pp.33-38 The asymptotic unbiasedness of S² in the linear regression model with moving average or particular s -th order autocorrelated disturbances Song, S. H.
  7. A course in probability theory Chung, K. L.
  8. Statistical Papers v.35 pp.28-36 Consistency, asymptotic unbiasedness and bounds on the bias of S² in the linear regression model with error component disturbances Baltagi, B.;Kramer, W.