ON EXTERNAL ANCILLARITY1
From MaRDI portal
In this paper the need for ancillary statistics which are not functions of the minimal sufficient statistic is stressed. Special attention is paid to models admitting a complete sufficient statistic, in which case ancillaries, if they exist, must be independent of the minimal sufficient statistic. Several classes of ancillary statistics are discussed, and it is shown that any ancillary is necessarily a function of ancillaries falling into these classes. The paper concludes with some applications.
Recommendations
Cited in
(15)- On the non-existence of ancillary statistics
- Sufficiency and ancillarity in characterization problems
- The reduction and ancillary classification of observations
- A condition for the nonexistence of ancillary statistics
- Simple ancillaries
- On rereading D. Basu's jointly sufficient statistic example made up of two ancillaries and miscellany
- APPLICATIONS AND EXTENSIONS OF BASU'S RESULTS ON MAXIMAL ANCILLARITY
- scientific article; zbMATH DE number 5511094 (Why is no real title available?)
- ANCILLARIES SUFFICIENT FOR THE SAMPLE SIZE1
- Mixtures, embedding, and ancillarity
- Pivotal ouantities based on the conditional distribution function transform in models with accessory parameters
- Some aspects of minimum variance unbiased estimation in presence of ancillary statistics
- scientific article; zbMATH DE number 218936 (Why is no real title available?)
- On the conditions for the existence of ancillary statistics in a curved exponential family
- scientific article; zbMATH DE number 4184622 (Why is no real title available?)
This page was built for publication: ON EXTERNAL ANCILLARITY1
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3704713)