3. The efficiency of the suggested estimators is found to be more superior
when the underlying distributions are symmetric as compared to
asymmetric.
4. The relative efficiency of SMRSS estimators to those estimators based on
SRS, SSRS and SRSS are increasing as the sample size increases. For
example, if the underlying distribution is normal with mean 0 and variance
1, the efficiency of SMRSS estimators relative to SSRS for n = 7 and 12,
are 3.7571 and 6.0638, respectively.