The Filtering Problem for Infinite Dimensional Stochastic Processes.
Abstract
The paper presents some recently obtained results on the nonlinear filtering problem for infinite dimensional processes. The optimal filter is obtained as the unique solution of certain measure valued equations. Robustness properties - both pathwise and statistical - are given and a preliminary result shows consistency with the stochastic calculus theory. Applications to random fields and models of voltage potential in neurophysiology are briefly discussed. Keywords: Markov processes; white noise.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1987
- Accession Number
- ADA186431
Entities
People
- G. Kallianpur
- R. L. Karandikar
Organizations
- University of North Carolina at Chapel Hill