Estimation problems for random fields from noisy data

From MaRDI portal





The author deals with the problem of the least mean-square linear estimation of the integral \[ A \xi =\int_{0}^{\infty} \int_{S_{n}}a(t,x)\xi(t,x) m_n( dx) dt \] which depends on the unknown values of a random field \(\xi(t,x),\) \(t\in R^1\), \(x\in S_{n}\), that is time-homogeneous and isotropic on a sphere \(S_{n}\) from observations of the field \(\xi(t,x)+\eta(t,x)\) for \(t<0\), \( x\in S_{n}\), where \( \eta(t,x)\) is a random field that is time-homogeneous and isotropic on a sphere \(S_{n}\) uncorrelated with \(\xi(t,x)\). Based on the spectral representation of time-homogeneous and isotropic on a sphere fields [see, for example, the author and \textit{M. I. Yadrenko}, Theory Probab. Math. Stat. 18, 115-124 (1979); translation from Teor. Veroyatn. Mat. Stat. 18, 106-115 (1978; Zbl 0409.62087) and ibid. 19, 129-139 (1980); resp. ibid. 19, 111-121 (1978; Zbl 0407.60059)] formulas are obtained for computing the value of the mean-square error and the spectral characteristic of the optimal linear estimate of the integral \(A \xi\). The least favorable spectral densities and the minimax (robust) spectral characteristics of the optimal estimates of the integral are determined for some classes of random fields. In the articles [Theory Probab. Math. Stat. 51, 137-146 (1995); translation from Teor. Jmovirn. Mat. Stat. 51, 131-139 (1994; Zbl 0939.60039) and ibid. 53, 137-148 (1996); resp. ibid. 53, 126-137 (1995; Zbl 0951.60056)] the author proposes more detailed analysis of minimax extrapolation problem from observations without noise which is based on the factorization of spectral densities.












This page was built for publication: Estimation problems for random fields from noisy data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4780940)