Significance tests and spurious correlation in regression models with autocorrelated errors

From MaRDI portal





In this paper, we report some results on the exact significance level when the usual F-statistic is used in a linear regression model with autocorrelated disturbances. The exact tail area probabilities sometimes differ substantially from the nominal size used in an 'F-test' and from upper-bound probabilities derived by \textit{J. Kiviet} [Bounds for the effects of ARMA disturbances on tests for regression coefficients. Tech. Rep., Univ. Amsterdam (1979)] which do not depend on the values of the regressors. A similar conclusion is also reached for the exact size of the significance tests for the spurious regressions considered by \textit{C. W. J. Granger} and \textit{P. Newbold} [Spurious regressions in econometrics. J. Econ. 2, 111-120 (1974) and Forecasting economic time series. Academic Press (1977)]. The results indicate once more that one has to be careful when using an algebraic F-test in the presence of autoregressive errors. However, then too, the Durbin-Watson test is expected to indicate the presence of autocorrelation.











This page was built for publication: Significance tests and spurious correlation in regression models with autocorrelated errors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3343264)