Search Filter
- Publications
- Abstract of Theses and Dissertations
- Database
- Sampling on Two Occasions When Strata Change in the Change Occasion
Sampling on Two Occasions When Strata Change in the Change Occasion
Abstract:
In this study, sampling on two occasions refers to sampling from the same conceptual population at two different and successive points of time. Sampling designs considered in this study are one- and two stage stratified sampling where simple random sampling without replacement is used in both occasions.
The study evaluated the effect of change in stratification classes on properties of estimates and derived estimators of means and the difference of means, variances of the estimators, and estimates of the variances. There were four cases of change in stratification considered. Case one, which may be considered the "ideal" situation, was when the strata in both occasions do not change. Case two was the situation wherein some strata in the first occasion were split into two or more new strata in the second occasion. The third case occurred when some strata in the first occasion were merged into some new strata in the second occasion.
The fourth case was the combination of cases two and three. In this case, the Best Linear Unbiased Estimator (BLUE) of means in the first and second occasions were derived.
Simulation was used to determine properties of estimates with regard to unbiasedness of the derived estimators of the estimates, and their variance estimates. Revenues of establishment for some five-digit PSIC industries in the Philippines for 1988 and 1989 were used in the simulation of one-stage stratified sampling on two occasions.
Results showed that the estimators in all cases were unbiased while the estimates were biased. As obtained from test data, the relative bias of estimate of change was numerically higher than the estimate of means. The lowest relative bias of estimate of change was about eight percent, while the highest was 28 percent. The lowest relative bias of estimates for the means was about 0.4 percent while the highest was seven percent. The simulation also showed that average of estimates for variances of means and difference ranged from 0.56 to 3 true variance.