论文信息 - Regression Design for Some Equivalence Classes of Kernels

Regression Design for Some Equivalence Classes of Kernels

Tn = tl < t2 < ... < t", ti e T} so that the variance Y2 of the Gauss-Markov estimate of 0 given {Y(t), t e T"} is as small as possible. This problem has been considered by Sacks and Ylvisaker, Wahba, and Hajek and Kimeldorf [3], [11], [12], [13], [14], [16] for various special cases of Q. It is known that a2T is bounded away from 0 as A = maxz jt+,-t4 tends to 0 if and only if f e Q, where Q is the unique reproducing kernel Hilbert space (RKHS) with reproducing kernel (RK)Q, see [8]. It will be assumed that the reader is familiar with the basic properties of RKHS as given in [8], [16], see also [1].

G. Wahba

[1] G. Kimeldorf,et al. Regression Designs in Autoregressive Stochastic Processes , 1974 .

[2] Grace Wahba,et al. A class of approximate solutions to linear operator equations , 1973 .

[3] G. Wahba. Convergence rates of certain approximate solutions to Fredholm integral equations of the first kind , 1973 .

[4] G. Wahba. On the Regression Design Problem of Sacks and Ylvisaker , 1971 .

[5] G. Wahba,et al. Some results on Tchebycheffian spline functions , 1971 .

[6] E. Parzen. STATISTICAL INFERENCE ON TIME SERIES BY RKHS METHODS. , 1970 .

[7] J. Sacks,et al. Designs for Regression Problems With Correlated Errors: Many Parameters , 1968 .

[8] Thomas Kailath,et al. On Measures Equivalent to Wiener Measure , 1967 .

[9] L. Shepp. Radon-Nikodym Derivatives of Gaussian Measures , 1966 .

[10] J. Sacks,et al. Designs for Regression Problems with Correlated Errors III , 1966 .

[11] M. Rosenblatt,et al. SINGULAR GAUSSIAN MEASURES IN DETECTION THEORY , 1963 .

[12] J. Hájek. On linear statistical problems in stochastic processes , 1962 .

[13] E. Parzen. An Approach to Time Series Analysis , 1961 .

[14] N. Aronszajn. Theory of Reproducing Kernels. , 1950 .