Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | SAM-P2.11 |
| Session: | Detection and Estimation |
| Time: | Tuesday, May 18, 15:30 - 17:30 |
| Presentation: |
Poster |
| Topic: |
Sensor Array and Multichannel Signal Processing: Signal detection and estimation |
| Title: |
MINIMAX REGRET ESTIMATION IN LINEAR MODELS |
| Authors: |
Yonina C. Eldar; Technion-Israel Institute of Technology | | |
| | Aharon Ben-Tal; Technion-Israel Institute of Technology | | |
| | Arkadi Nemirovski; Technion-Israel Institute of Technology | | |
| Abstract: |
We develop a new linear estimator for estimating an unknown vector x in a linear model, in the presence of bounded datauncertainties. The estimator is designed to minimize theworst-case regret across all bounded data vectors, namelythe worst-case difference between the MSE attainable using alinear estimator that does not know the true parameters x, andthe optimal MSE attained using a linear estimator that knowsx. We demonstrate through several examples that the minimaxregret estimator can significantly increase the performance overthe conventional least-squares estimator, as well as several other least-squares alternatives. |
| |
| Back | |