SEQUENTIAL SAMPLING OF NORMAL AND NONNORMAL POPULATIONS

This paper reports a simulation-based exploration into the computation of point and interval estimates for data arising from sequential samp ling of populations with a variety of underlying distributions. We con clude that the coverage probability of the standard frequentist confid ence interval estimates is overstated. The effect of non-normal behavi or in the underlying population upon the properties of the interval es timate varies, depending upon the severity and type of anomaly (skewne ss, kurtosis, etc). Sequential sampling of certain nonsymmetric distri butions performed very poorly compared with simple random sampling. We assess other interval estimates that do not overstate the coverage pr obability if the underlying population is normal (Gaussian).