COMPARISON OF k-CLASS ESTIMATORS WHEN THE DISTURBANCES ARE SMALL.
STANFORD UNIV CALIF DEPT OF STATISTICS
Pagination or Media Count:
Asymptotic expansions of the variance of the disturbance around zero are found for the bias to order sigma squared and moment matrix to order sigma to the 4th power of all k-class estimators, including limited information maximum likelihood. The discussion relates the results to large sample asymptotic theory, Monte Carlo experiments and exact fixed-sample studies of linear simultaneous equation regression techniques. Several uniformities of interest are found. Author
- Economics and Cost Analysis
- Statistics and Probability