Inverse Bernoulli Trials with Dependence.
Technical summary rept.,
WISCONSIN UNIV MADISON MATHEMATICS RESEARCH CENTER
Pagination or Media Count:
A model for inverse sampling with dependence between successive Bernoulli random variables is presented. Sufficient statistics for the model are given and their joint distribution worked out from the sampling theory of a modified geometric distribution. Simple estimators of the parameters are recommended and shown to be efficient in large samples. A uniformly most powerful test of independence is presented along with an application to some virus research data. Author
- Statistics and Probability