Monte Carlo experiments establish that the usual ``t-statistic'' used fortesting for first-order serial correlation with artificial regressions is far from being distributed as a Student's t in small samples. Rather, it is badly biased in both mean and variance and results in grossly misleading...

The traditional two-step procedure for correcting for heteroskedasticity uses a consistent but biased estimator for the variances $\bfg\sigma_t^2$ in enacting the second step. An estimator is developed here that is unbiased in the presence of heteroskedasticity. Its behavior is examined along...

This working paper provides some preliminary results on the computational feasibility of nonlinear full information maximum likelihood (NECML) estimation. Severa1 of the test cases presented were also subjected to nonlinear three stage least square (NLBSLS) estimation in order to illustrate the...