Unbiased estimation of the population variance using Midzuno-Sen type sampling scheme

We have employed ms type sampling scheme to propose two unbiased strategies
for estimating the population variance. These strategies have been compared with
certain known ones, and necessary and sufficient conditions have been obtained
for their superior performance as compared to the known ones. An unbiased
variance estimator of the population variance has also been worked out. Real-life
data are shown to yield substantial gains via these strategies.