Standard

Exact confidence regions for linear regression parameter under external arbitrary noise. / Senov, A.; Amelin, K.; Amelina, N.; Granichin, O.

In: Proc. of the 2014 American Control Conference (ACC),. 2014. p. 5097-5102.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research

Harvard

Senov, A, Amelin, K, Amelina, N & Granichin, O 2014, Exact confidence regions for linear regression parameter under external arbitrary noise. in In: Proc. of the 2014 American Control Conference (ACC),. pp. 5097-5102. https://doi.org/10.1109/ACC.2014.6859436

APA

Senov, A., Amelin, K., Amelina, N., & Granichin, O. (2014). Exact confidence regions for linear regression parameter under external arbitrary noise. In In: Proc. of the 2014 American Control Conference (ACC), (pp. 5097-5102) https://doi.org/10.1109/ACC.2014.6859436

Vancouver

Senov A, Amelin K, Amelina N, Granichin O. Exact confidence regions for linear regression parameter under external arbitrary noise. In In: Proc. of the 2014 American Control Conference (ACC),. 2014. p. 5097-5102 https://doi.org/10.1109/ACC.2014.6859436

Author

Senov, A. ; Amelin, K. ; Amelina, N. ; Granichin, O. / Exact confidence regions for linear regression parameter under external arbitrary noise. In: Proc. of the 2014 American Control Conference (ACC),. 2014. pp. 5097-5102

BibTeX

@inproceedings{22bdff70bb1a43ec86d43794369dcd91,
title = "Exact confidence regions for linear regression parameter under external arbitrary noise",
abstract = "The paper propose new method for identifying non-asymptotic confidence regions for linear regression parameter under external arbitrary noise. This method called Modified Sign-Perturbed Sums (MSPS) method and it is a modification of previously proposed one, called Sign-Perturbed Sums which is applicable only in case of symmetrical centred noise. MSPS algorithm correctness and obtained confidence region convergence are proved theoretically under some additional assumptions. SPS and MSPS methods are compared basing on simulated data. Few advantages of MSPS method in case of biased and asymmetric noise are illustrated.",
keywords = "Linear systems, Randomized algorithms, Uncertain systems",
author = "A. Senov and K. Amelin and N. Amelina and O. Granichin",
year = "2014",
doi = "10.1109/ACC.2014.6859436",
language = "English",
isbn = "9781479932726",
pages = "5097--5102",
booktitle = "In: Proc. of the 2014 American Control Conference (ACC),",

}

RIS

TY - GEN

T1 - Exact confidence regions for linear regression parameter under external arbitrary noise

AU - Senov, A.

AU - Amelin, K.

AU - Amelina, N.

AU - Granichin, O.

PY - 2014

Y1 - 2014

N2 - The paper propose new method for identifying non-asymptotic confidence regions for linear regression parameter under external arbitrary noise. This method called Modified Sign-Perturbed Sums (MSPS) method and it is a modification of previously proposed one, called Sign-Perturbed Sums which is applicable only in case of symmetrical centred noise. MSPS algorithm correctness and obtained confidence region convergence are proved theoretically under some additional assumptions. SPS and MSPS methods are compared basing on simulated data. Few advantages of MSPS method in case of biased and asymmetric noise are illustrated.

AB - The paper propose new method for identifying non-asymptotic confidence regions for linear regression parameter under external arbitrary noise. This method called Modified Sign-Perturbed Sums (MSPS) method and it is a modification of previously proposed one, called Sign-Perturbed Sums which is applicable only in case of symmetrical centred noise. MSPS algorithm correctness and obtained confidence region convergence are proved theoretically under some additional assumptions. SPS and MSPS methods are compared basing on simulated data. Few advantages of MSPS method in case of biased and asymmetric noise are illustrated.

KW - Linear systems

KW - Randomized algorithms

KW - Uncertain systems

U2 - 10.1109/ACC.2014.6859436

DO - 10.1109/ACC.2014.6859436

M3 - Conference contribution

SN - 9781479932726

SP - 5097

EP - 5102

BT - In: Proc. of the 2014 American Control Conference (ACC),

ER -

ID: 7006291