Correlogram estimation of response functions of linear systems in scheme of some independent samples

I. P. Blazhievska

Teoriâ slučajnyh processov, Tome 17 (2011) no. 1, pp. 16-27.

Résumé

The problem of estimation of an unknown response function from $L_{2}(\mathbf{R})$ of a linear system is considered. The inputs are supposed to be stationary zero-mean Gaussian almost surely sample continuous processes. We take the integral-type sample input-output cross-correlograms as estimators of the response function and apply the scheme of some independent samples, when the pair of inputs and outputs are observed. The asymptotic normality of the distributions of centered cross-correlogram estimations in the space of continuous functions and the construction of the confidence bands for the limiting process are discussed.

Détail

Keywords: Response function, sample continuity, sample cross-correlogram, scheme of some independent samples, asymptotic normality.

Bibliographie

[1] J. S. Bendat, A. G. Piersol, Engineering Applications of Correlation and Spectral Analysis, Wiley, New York, 1993

[2] I. P. Blazhievska, “On asymptotic behavior of the error term in cross-correlogram estimation of response functions in linear systems”, Theory of Stochastic Processes, 16 (32):2 (2010), 5–11

[3] V. V. Buldygin, “On some comparison inequality for the distribution of the maximum of the Gaussian process”, Theor. Probab. and Math. Statist., 28 (1983), 9–14 (in Russian)

[4] V. V. Buldygin, “Limiting theorems in functional spaces and one problem in statistics of stochastic processes”, Probabilistic Methods of Infinite-Dimensional Analysis, Inst. of Math. AN USSR, Kiev, 1980, 24–36 (in Russian)

[5] V. V. Buldygin, I. P. Blazhievska, “On asymptotic behavior of cross-correlogram estimators of response functions in linear Volterra systems”, Theory of Stochastic Processes, 15 (31):2 (2009), 84–98

[6] V. V. Buldygin, E. V. Ilarionov, “On some problem in statistics of stochastic fields”, Probabilistic Infinite-Dimensional Analysis, Inst. of Math. AN USSR, Kiev, 1981, 6–14 (in Russian)

[7] V. V. Buldygin, Yu. V. Kozachenko, Metric Characterization of Random Variables and Random Processes, Amer. Math. Soc., Providence, RI, 2000

[8] V. V. Buldygin, V. G. Kurotschka, “On cross-correlogram estimators of the response function in continuous linear systems from discrete observations”, Random Oper. and Stoch. Eq., 7:1 (1999), 71–90

[9] V. V. Buldygin, Fu Li, “On asymptotical normality of an estimation of unit impulse responses of linear systems I, II”, Theor. Probab. and Math. Statist., 54, 55 (1997), 17–24; 29–36

[10] V. Buldygin, F. Utzet, V. Zaiats, “Asymptotic normality of cross-correlongram estimates of the response function”, Statistical Interference for Stochastic Processes, 7 (2004), 1–34

[11] V. V. Buldygin, A. B. Charazishvili, The Brunn–Minkowski Inequality and Its Applications, Naukova Dumka, Kiev, 1985 (in Russian)

[12] R. M. Dudley, “Sample functions of the Gaussian process”, Ann. of Probab., 1 (1973), 66–103

[13] K. Fernigue, “Regularite des trajectoires des fonctions aleatoires gaussiennes”, Lect. Notes in Math., 450, Springer, Berlin, 1975, 1–96

[14] U. Grenander, Probabilities on Algebraic Structures, Wiley, New York, 1963

[15] N. C. Jain, M. B. Marcus, “Central limit theorem for CS-valued ramdom variables”, J. Funct. Analysis, 19:3 (1975), 216–231

[16] H. Cramér, M. Leadbetter, Stationary and Related Stochastic Processes: Sample Function Properties and Their Applications, Wiley, New York, 1967

[17] N. N. Leonenko, A. V. Ivanov, Statistical Analysis of Stochastic Fields, Vyshcha Shkola, Kiev, 1986 (in Russian)

[18] M. B. Marcus, L. A. Shepp, “Countinuity of Gaussian processes”, Trans. Amer. Math. Soc., 151 (1970), 377–391