On Information and Suuciency Abbreviated Title: Information and Suuciency
نویسنده
چکیده
The information deviation between any two nite measures cannot be increased by any statistical operations (Markov morphisms). It is invariant if and only if the morphism is suucient for these two measures. AMS 1991 Classiication: Primary 62B10 (stat info); Secondary 62B20 (meas th), 62B05 (suu stat), 94A17 (meas of info). Acknowledgment: I thank Kim Sau Chung for pointing out several inaccuracies in a previous version.
منابع مشابه
On the extended linear complementarity problem
For the extended linear complementarity problem 11], we introduce and characterize column-suuciency, row-suuciency, and P-properties. These properties are then specialized to the vertical , horizontal, and mixed linear complementarity problems.
متن کاملOn the Spectrum of the Linear
The continuous spectrum of the linear Boltzmann operator with constant eld is derived. It is found that a suuciency relation for runaway phenomena is consistent with another suuciency relation for the hydrody-namic regime to exist. There is a further class of systems whose behaviour lies in between these two extremes.
متن کاملAxiomatic approach to feature subset selection basedon
In this paper an axiomatic characterisation of feature subset selection is presented. Two axioms are presented: suuciency axiom | preservation of learning information, and necessity axiom | minimising encoding length. The suuciency axiom concerns the existing dataset and is derived based on the following understanding: any selected feature subset should be able to describe the training dataset ...
متن کاملSuucient Markov Chains
We develop a theory of suuciency for Markov chains. In particular, we prove that to analyse convergence in total variation distance, it suuces to obtain bounds on a suucient chain.
متن کاملProbabilities of causation : Three counterfactualinterpretations and their identi cationJudea
According to common judicial standard, judgment in favor of plaintii should be made if and only if it is \more probable than not" that the defendant's action was the cause for the plaintii's damage (or death). This paper provides formal semantics, based on structural models of counterfactuals, for the probability that event x was a necessary or suucient cause (or both) of another event y. The p...
متن کامل