A Monte Carlo comparison of regression estimators in the presence of autocorrelation and collinearity
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 r...
| Main Author: | |
|---|---|
| Language: | English |
| Published: |
2010
|
| Online Access: | http://hdl.handle.net/2429/23555 |