Kalman Filter Models for Extrapolations in Dose-Response Experiments and Accelerated Life-Tests.
Abstract
Kalman-Filter models with Gaussian innovations provide a useful and easy to implement tool for inference from dose-response experiments and accelerated life-tests. Their main advantage stems from the fact that the system equation of such models allows for the uncertainty and possible changes in a proposed dose-response relationship. This is in contrast to the currently used approaches wherein there is an implicit commitment to the validity of an assumed relationship. In this paper, we overview our recent work in the above general area and suggest avenues for future research. (AN)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 09, 1988
- Accession Number
- ADA293949
Entities
People
- Nozer Singpurwalla
Organizations
- George Washington University