Synopses & Reviews
This monograph develops an approach to statistical inference that is both comprehensive in its treatment of statistical principles and sufficiently powerful to be applicable to a variety of important practical problems, such as inference for stochastic processes and classes of estimating functions. Some of the consequences of extending standard concepts of ancillarity, sufficiency and completeness are examined in this setting. The development is mathematically mature in its use of Hilbert space methods, but not mathematically difficult. Thus, the construction of this theory is rich in statistical tools for inference without the difficulties found in modern developments, such as likelihood analysis of stochastic processes or higher order methods.
Table of Contents
Contents: Introduction.- The Space of Inference Functions: Ancillarity, Sufficiency and Projection.- Selecting an Inference Function for 1-Parameter Models.- Nuisance Parameters.- Inference Under Restrictions.- Inference for Stochastic Processes.- References.- Index.