For system identification problems with stochastic noise, maximum likelihood estimators are frequently used. If noise is deterministic, worst case optimal algorithms should be considered. In this paper we study the following problem: Under what circumstances are maximum likelihood estimators optimal in the worst case?
|Number of pages||6|
|Journal||Systems and Control Letters|
|State||Published - Apr 1988|
Bibliographical noteFunding Information:
* This research was developed while Dr. Tempo was visiting the Department of Computer Science, Columbia University supported by a NATO-CNR advanced fellowship and by funds of Ministero della Pubblica Istruzione. ** Supported in part by the National Science Foundation under Grant DCR-86-03674.
- Maximum likelihood estimators
- Noisy information
- Optimal algorithms
- System identification
- Worst case
ASJC Scopus subject areas
- Control and Systems Engineering
- Computer Science (all)
- Mechanical Engineering
- Electrical and Electronic Engineering