Carè, Algo and Csáji, Balázs Csanád and Campi, Marco (2016) SignPerturbed Sums (SPS) with Asymmetric Noise: Robustness Analysis and Robustification Techniques. In: 55th IEEE Conference on Decision and Control, December 1214, 2016, Las Vegas, Nevada.

Text
RobustSPSIEEECDC2016.pdf  Updated Version Download (351kB)  Preview 
Abstract
SignPerturbed Sums (SPS) is a recently developed finite sample system identification method that can build exact confidence regions for linear regression problems under mild statistical assumptions. The regions are wellshaped, e.g., they are centred around the leastsquares (LS) estimate, starconvex and strongly consistent. One of the main assumptions of SPS is that the distribution of the noise terms are symmetric about zero. This paper analyses how robust SPS is with respect to the violation of this assumption and how it could be robustified with respect to nonsymmetric noises. First, some alternative solutions are overviewed, then a robustness analysis is performed resulting in a robustified version of SPS. We also suggest a modification of SPS, called LADSPS, which builds exact confidence regions around the leastabsolute deviation (LAD) estimate instead of the LS estimate. LADSPS requires less assumptions as the noise needs only to have a conditionally zero median (w.r.t. the past). Furthermore, that approach can also be robustified using similar ideas as in the LSSPS case. Finally, some numerical experiments are presented.
Item Type:  Conference or Workshop Item (Paper) 

Subjects:  Q Science / természettudomány > QA Mathematics / matematika > QA75 Electronic computers. Computer science / számítástechnika, számítógéptudomány T Technology / alkalmazott, műszaki tudományok > TK Electrical engineering. Electronics Nuclear engineering / elektrotechnika, elektronika, atomtechnika 
Depositing User:  Dr. Balázs Csanád Csáji 
Date Deposited:  05 Oct 2018 06:37 
Last Modified:  05 Oct 2018 06:37 
URI:  http://real.mtak.hu/id/eprint/86582 
Actions (login required)
Edit Item 