No label defined (Q4258796)

The problem of minimax-robust linear least-squares estimation of a linear functional \(A\xi\) that depends on the unknown values of a weakly stationary stochastic process \(\xi(t)\), \(t\in R^1\), from observations of the process \(\xi(t)+\eta(t)\), \(t\in R^1\setminus[0,T]\), is considered. Formulas are obtained for computing the value of the mean-square error and the spectral characteristic of the optimal linear estimate of the functional \(A\xi\). It is assumed that the spectral densities \(f(\lambda)\) and \(g(\lambda)\), the signal process \(\xi(t)\) and the noise process \(\eta(t)\) are not known exactly. But it is assumed that \(f(\lambda)\in D_{f}\) and \(g(\lambda)\in D_{g}\), where \(D_{f}\) and \(D_{g}\) are convex sets of spectral densities. The least favorable spectral densities \(f_{0}(\lambda)\in D_{f}\), \(g_{0}(\lambda)\in D_{g}\) and minimax (robust) spectral characteristics of the optimal estimates of the functional \(A\xi\) are determined for some special classes of spectral densities. For the entire collection see [Zbl 0896.00028].