Consistency of the Correlogram Estimator of Covariance Function of Random Noise in Nonlinear Regression
Keywords:Nonlinear regression model, Stationary Gaussian noise, Covariance function, Residual correlogram estimator, Consistency, Pseudometric, Metric massiveness, Metric entropy
Background. For nonlinear regression model with continuous time and mean square continuous separable measurable Gaussian stationary random noise with zero mean and square-integrable spectral density the problem of statistical estimation of unknown covariance function of random noise in the presence of regression function nuisance parameter is considered.
Objective. To study asymptotic behavior of residual correlogram estimator of random noise covariance function using consistent least squares estimator of regression function unknown parameter.
Methods. To obtain the paper results we rely on the use of methodological machinery of monographs by V.V. Buldygin and Yu.V. Kozachenko (2000), specifically, the theory of quadratically Gaussian random processes, and A.V. Ivanov and N.N. Leonenko (1989).
Results. Sufficient conditions of weak consistency in unknown metric of residual correlogram estimator are obtained. The rate of covariance to zero of the strong consistency in unknown metric of the estimator is formulated.Conclusions. The results obtained given opportunity to continue research of asymptotic properties of Gaussian stationary random noise covariance function residual correlogram estimator in nonlinear regression model and to prove a functional theorem in the space of continuous function on asymptotic confidence intervals for unknown covariance function of random noise.
O.V. Ivanov and K.K. Moskvychova, “Stochastic asymptotic expansion of random noise covariance function сorrelogram estimator in nonlinear regression model”, Teoria Imovirnostey i Matematychna Statystyka, no. 90, pp. 78–91, 2014 (in Ukrainian).
O.V. Ivanov and K.K. Moskvychova, “Asymptotic expansion of the moments of random noise covariance function сorrelogram estimator in nonlinear regression model”, Ukrainsky Matematychny Zhurnal, vol. 66, no. 6, pp. 787–805, 2014 (in Ukrainian).
A.V. Ivanov and N.N. Leonenko, Statistical Analysis of Random Fields. Dordecht, Boston, London: Kluwer Academic Publishers, 1989, 244 p.
V.V. Buldygin and Yu.V. Kozachenko, Metric Characterization of Random Variables and Random Processes (Transl. of Math. Monographs). Am. Math. Soc., 2000, 257 p.
I.K. Matsak, Elements of the Theory of Extreme Values.Kyiv,Ukraine: Comprint, 2014, 210 p. (in Ukrainian).
N.N. Leonenko, Limit Theorems for Random Fields with Singular Spectrum. Dordecht, Boston, London: Kluwer Academic Publishers, 1999, 401 p.
A.V. Ivanov, Asymtotic Theory of Nonlinear Regression. Dordecht, Boston, London: Kluwer Academic Publishers, 1997, 327 p.
LicenseCopyright (c) 2017 NTUU KPI Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under CC BY 4.0 that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work