Article Abstract:
An extended matrix inverse result has been developed with a new unified double length regression method to calculate the two-step generalized least squares (2SGLS) estimators of two types of rational expectations models. This newly-developed double-length regression computational method is limited to certain types of rational expectations models and is not suitable for models with MA error and heteroscedasticity. Modified variations of 2SGLS estimators and their applications were also discussed.
User Contributions:
Comment about this article or add new information about this topic:
Article Abstract:
Specification testing has become critical in gauging the robustness of proposed empirical econometric models. A study was therefore conducted to compare the performance of different tests for functional form and heteroscedasticity for single-equation instrumental variable (IV) regressions. It illustrates the significance of using forecast values of the endogenous variables in the development of functional form and heteroscedasticity tests for IV regressions.
User Contributions:
Comment about this article or add new information about this topic:
Article Abstract:
A new unified double length regression method based on an extension of a matrix inverse result of Higgins (1994) is proposed. The method is used to compute for the two-step generalized least squares estimators of rational expectations models with current anticipated and unanticipated element. Estimators can be directly applied to a majority of standard econometric software.
User Contributions:
Comment about this article or add new information about this topic: