Institute of Information Theory and Automation

You are here

Some Results and Problems in the Theory of Fisher Information

Date: 
2015-11-23 14:00
Room: 
Name of External Lecturer: 
Abram Kagan
Affiliation of External Lecturer: 
Dept. of Mathematics, Univ. of Maryland, College Park
After presenting general properties of the Fisher information, not so well known results in the signal plus noise model (Stam inequality, Carlen's superadditivity, contamination, small perturbations) will be discussed.

Close connections to parameter estimation, especially to the Pitman estimators will be demonstrated as well as properties of the latter.

Lower bounds for the Fisher information and the role of special distributions will be shown.

kratochvil: 2015-11-02 11:09