A note on the use of logged variables in interaction terms: why units matter
Units of measurement for models with logged variables must be chosen with care to avoid presenting misleading information. For loglinear models the coefficient of the logged variable hinges on another variable, which leads to inferences that are scale dependent. In cointegration tests, testing unit root integration order is not applicable for models with at least one variable that is logarithmized.
Publication Name: Oxford Bulletin of Economics & Statistics
A Comparison of the Amemiya GLS and the Lee- Maddala-Trost GZSLS in a Simultaneous-Equations Tobit Model
Performance comparison is made of a class of general least square estimators. The Amemiya GLS estimators are asymptotically more efficient. The Amemiya estimator is statistically burdensome.
Publication Name: Journal of Econometrics
- Abstracts: Earnings management and the performance of seasoned equity offerings. The exercise and valuation of executive stock options
- Abstracts: Costly verification of cost performance and the competition for incentive contracts. Information sharing in oligopoly: the truth-telling problem
- Abstracts: On the relevance of learning and evolution to economic theory. Royal Commission 'Report on Transport and the Environment' - economic effects of recommendations
- Abstracts: Improvements to CPI procedures: sample rotation of the used car and truck sample. Extending the improvement in CPI sample rotation procedures
- Abstracts: The effect of changes in ownership structure on performance: evidence from the thrift industry. Sources of gains in horizontal mergers: evidence from customer, supplier, and rival firms