Abstract
The usefulness of stochastic optimization for sample allocation in stratified sampling is studied. Three models of stochastic optimization are compared: E-Model, Modified E-model and V-model, recently presented by Diaz-Garcia and Garay-Tapia (Comput. Statistics Data Anal., 3016–3026, 51, 2007), with the classical sample allocation, which distributes the costs among strata in such a way that the variance of an estimator is minimized. To make the comparison, a simulation study was conducted. None of the methods was the most efficient for all cases, but usually the classical allocation was the most efficient, followed by the E-model, quite similar to the former.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
