UBC Theses and Dissertations
A Monte Carlo comparison of regression estimators in the presence of autocorrelation and collinearity Gosling, Barbara J.
In time series regression modelling, first-order autocorrelated errors are often a problem. When the data also suffers from collinear independent variables, generalized least squares estimation is no longer the best alternative to ordinary least squares. The Monte Carlo simulation illustrates that ridge estimation using data transformed according to the generalized least squares method provides estimates of the regression coefficients which are superior to generalized least squares, ridge and ordinary least squares estimates. The analysis of a set of 'typical' econometric data further supports the application of this method referred to as generalized ridge when both autocorrelation and collinearity are present.
Item Citations and Data