A ROBUST LEAST TRIMMEDD SQUARES FOR AUTOCORRELATED RESIDUALS

Abstract

It is well known that, the classical Durbin-Watson test is the most commonly used regression technique for detecting autocorrelation. However, this test is affected by outliers. Therefore, we cannot both detect the autocorrelation of disturbances and remedy the harm caused by this phenomenon. In this paper, we conjecture how to robustify the Durbin-Watson test for detecting the autocorrelation problem. A description of the least trimmed squares regression and least weighted squares follows . Thus, we can to robustify the Durbin-Watson test for residuals of the least trimmed squares regression. An example with real data supports the practical character of this paper

Keywords: Cochrane-Orcutt transformation, Durbin-Watson test, Durbin-Watson test in robust regression., Least Trimmed Squares, Reweighted Least Squares

Article Review Status: Published

Pages: 1-12 (Download PDF)

Creative Commons Licence
This work by European American Journals is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License

  • Our Journal Publishing Partners