Journal of Official Statistics, Vol.16, No.1, 2000. pp. 31–37
A Note on the Use of Inverse Sampling: Point Estimation Between Successive Infections
Abstract:This article considers point estimation of the effect of the primary infection on the likelihood of developing a secondary infection. The article notes that the use of inverse sampling can avoid the theoretical limitations, such as bias or nonexistence of variance, in application of the maximum likelihood estimator under the commonly-assumed multinomial sampling. Under inverse sampling, this article derives the uniformly minimum variance unbiased estimators (UMVUEs) of the risk ratio and risk difference, as well as their corresponding variances. This article further develops the UMVUEs of these variances as well. Finally, the article includes a discussion on interval estimation and applies Monte Carlo simulation to evaluate and compare the performance of the interval estimators that are derived from the point and variance estimators obtained here.
Keywords:Point estimation; uniformly minimum variance unbiased estimator; maximum likelihood estimator; inverse sampling; risk ratio; risk difference.
Copyright © Statistics Sweden 1996-2018. Open AccessISSN 0282-423XCreated and Maintained by OKS Group