Comparison Of Four Estimators Under Sampling Without Replacement.

Abstract

In some applications, it is cost efficient to sample data in more than one stage. I the first stage, a random sample is drawn then stratisfied accordingly to some easily measured attributes. This project described four estimator's for the treatment of samples drawn without replacement when equal and unequal probabilities are considered. It also comparer their resulting standard error under sampling without replacement using data on diabetic patients in Niger state do the years 2000, 2001, 2002 and 2003. The results were obtained using a programme written in Microsoft Visual C++ programming language. It was observed that the two-stage sampling under unequal probabilities without replacement is always better than the other three estimator's considered.