On Taking Sequential Decisions in a Changing Environment.
SACLANT ASW RESEARCH CENTRE LA SPEZIA (ITALY)
Pagination or Media Count:
An attempt is made to extend Walds sequential decision theory to the case where the state of the system being observed can change during the observation-decision process. Most of the results obtained concern systems that can have only two possible states and are based on a partial differential equation which describes approximately the evolution in time of the probability ratio or likelihood ratio. In particular, the following elements can be deduced from that partial differential equation for a sequential decision procedure with constant thresholds mean number of observations before a terminal decision is taken, probabilities of errors, and expected loss. Author
- Operations Research