Interval Estimation of Prescribed Accuracy
For a fixed sample problem the accuracy of an estimator of a parameter θ typically depends on the unknown value of θ and perhaps also on the value of an unknown nuisance parameter λ. To achieve estimators of prescribed accuracy, it seems natural to proceed sequentially, to get some preliminary idea of the value of (θ,λ) and use the preliniminary information to determine the final sample size.
KeywordsCoverage Probability Interval Estimation Sequential Probability Ratio Test Prescribe Accuracy Approximate Confidence Interval
Unable to display preview. Download preview PDF.