Semi-parametric modification of cumulative sum algorithms for the change-point detection of non-Gaussian sequences

• Autorzy: Zabolotnii S. W , Warsza Z. L.
• Języki publikacji: EN

Abstrakty

EN

The expansion of logarithm likelihood ratio in the stochastic series to find the sequential change-point detection of non-Gaussian sequences is used. The moment criteria of the minimum of upper limit error probabilities sum to find the expansion coefficients is applied. The proposed method is a semi-parametric type of cumulative sum (CUSUM) algorithm which needs of higher-order statistics. Results show that polynomial algorithms are more effective in comparison with similar non-parametric procedures.

Słowa kluczowe

EN

change point; CUSUM algorithm; non-Gaussian sequence; stochastic polynomial; high order statistics

• Wydawca: Wydawnictwo PAK
• Czasopismo: Measurement Automation Monitoring
• Rocznik: 2015
• Tom: Vol. 61, No. 12
• Strony: 532--534
• Opis fizyczny: Bibliogr. 9 poz., rys., wzory

Twórcy

autor

Zabolotnii S. W

• Cherkasy State University of Technology, Cherkasy, Ukraine

autor

Warsza Z. L.

• Research Institute of Automation and Measurements PIAP, 202 Jerozolimskie Ave., 02-486 Warsaw, Poland

Bibliografia

• [1] Zabolotnii S. W., and Warsza Z. L.: Semi-parametric estimation of the change-point of mean value of non-Gaussian random sequences by polynomial maximization method. Proceedings of 13th IMEKO TC10 Workshop on Technical Diagnostics. Warsaw UT, pp. 189-194, 2014.
• [2] Basseville M., and Nikiforov I. V.: Detection of abrupt changes: theory and application. Prentice-Hall, New Jork USA, 1993.
• [3] Brodsky B., and Darkhovsky B.: Nonparametric Methods in change-point problems. Kluwer Academic Publishers, Dordrecht, the Netherlands, 1993.
• [4] Lokajicek T., and Klima K.: A first arrival identification system of acoustic emission (AE) signals by means of a higher-order statistics approach. Measurement Science and Technology, vol. 17, pp. 2461-66, 2006.
• [5] Yih-Ru Wang: The signal change-point detection using the high-order statistics of log-likelihood difference functions. Proccedings of Acoustics, Speech and Signal Processing, ICASSP IEEE International Conf. pp. 4381-4384, 2008.
• [6] Hilas C. S., Rekanos I. T., and Mastorocostas P. A.: Change point detection in time series using higher-order statistics: a heuristic approach. Math. Problems in Engineering, ID 317613, 2013.
• [7] Kunchenko Y.: Polynomial Parameter Estimates of Close to Gaussian Random variables. Aachen, Shaker Verlag, 2002.
• [8] Kunchenko Y. P.: A moment performance criterion of a decision making for testing simple statistical hypothesis. IEEE, International Symposium on Information Theory, Ulm, Germany, p. 40, 1997.
• [9] Page E. S.: A test for a change in a parameter occuring at an unknown point. Biometrika, vol. 42, pp. 523-527, 1955.