Information inequalities for the minimax risk of sequential estimators (with applications)
Volume 25 / 1998
Applicationes Mathematicae 25 (1998), 85-100
DOI: 10.4064/am-25-1-85-100
Abstract
Information inequalities for the minimax risk of sequential estimators are derived in the case where the loss is measured by the squared error of estimation plus a linear functional of the number of observations. The results are applied to construct minimax sequential estimators of: the failure rate in an exponential model with censored data, the expected proportion of uncensored observations in the proportional hazards model, the odds ratio in a binomial distribution and the expectation of exponential type random variables.