New transformed estimators in stratified random sampling: A case study on rubber production in Thailand.

Estimating the rubber production in Thailand, the world’s leading rubber supplier, can help the Thai government to prepare for rubber cultivation in policy planning. A transformation technique can be used to improve the efficiency of estimating the average rubber yield by reducing the biases and mean square error. A group of population mean estimators has been suggested under stratified random sampling utilizing a transformed auxiliary variable. The biases and mean square errors of the proposed estimators are investigated. Simulation studies and an application to rubber production data in Thailand have been applied to assess their performances under stratified random sampling where the yield of rubber varies depending upon the region. The results show that the estimates of rubber yields with the proposed estimators had small biases and mean square errors. The best estimator gave an estimated rubber production of 1,140 kilogram/hectare, which is close to the population mean of the yields of rubber [ABSTRACT FROM AUTHOR]