Professor Christos Agiakloglou
22nd March 2018 at 13.00 pm
Mithras House, University of Brighton
When time series data is used in econometrics serially correlated errors are most likely to appear. Autocorrelation will also be detected in regression analysis as an indication of a false specification between two variables. This study examines the problem of serially correlated errors in the context of spurious regression showing evidence of removing the presence of this phenomenon both theoretically as well as empirically by applying the Cochrane-Orcutt procedure.