Abstract
Suppose that ¯(x1),...,¯Z(xn). are observations of vector-valued random function ¯(x). In the isotropic situation, the sample variogram γ*(h) for a given lag h is
where s(h) is a set of paired points with distance h and N(h) is the number of pairs in s(h)..
For a selection of lags h1, h2, .... hk such that N (h1) > O. we obtain a ktuple of (semi) positive definite matrices\(\bar \gamma ^ * (h_{ 1} ),. . . ., \bar \gamma ^ * (h_{ k} )\). We want to determine an orthonormal matrix B which simultaneously diagonalizes the\(\bar \gamma ^ * (h_{ 1} ),. . . ., \bar \gamma ^ * (h_{ k} )\) or nearly diagonalizes them in the sense that the sum of squares of offdiagonal elements is small compared to the sum of squares of diagonal elements. If such a B exists, we linearly transform\(\overline Z (x)\) by\(\overline Y (x) = B\overline Z (x)\). Then, the resulting vector function\(\overline Y (x)\) has less spatial correlation among its components than\(\overline Z (x)\) does. The components of\(\overline Y (x)\) with little contribution to the variogram structure may be dropped, and small crossvariograms fitted by straightlines. Variogram models obtained by this scheme preserve the negative definiteness property of variograms (in the matrix-valued function sense). A simplified analysis and computation in cokriging can be carried out. The principles of this scheme arc presented in this paper.
Similar content being viewed by others
References
Clarkson, D. B., 1988a, A remark on algorithm AS 211, the F-G diagonalization algorithm: Applied Statistics, v. 37, no. 1, p. 147–151.
Clarkson, D. B., 1988b, A least squares version of Algorithm AS 211, the F-G diagonalization algorithm: Applied Statistics, v. 37, no. 2, p. 317–321.
Cressie, N., 1985, Fitting variogram models by weighted least squares: Math. Geology, v. 17, no. 5, p. 563–586.
Cressie, N., 1991, Statistics for spatial data: Wiley-Interscience, New York, 900 p.
De Leeuw, J., and Pruzansky, S., 1978, A new computation method to fit the weighted Euclidean distance model: Psychometrika, v. 43, no. 4, p. 479–490.
Flury, B. N., and Constantine, G., 1985, The FG-diagonalization algorithm: Appl. Statist., v. 35, no. 2, p. 177–183.
Flury, B. N., and Gautschi, W., 1986, An algorithm for simultaneous orthogonal transformation of several positive definite symmetric matrices to nearly diagonal form: SIAM Jour. Sci. Stat. Comput., v. 7, no. 1, p. 169–184.
Goovaerts, P., 1994, On a controversial method for modeling a coregionalization: Math. Geology, v. 26, no. 2, p. 197–204.
Myers, D. E., 1982, Cokriging: the matrix form: Math. Geology, v. 14, no. 3, p. 249–257.
Myers, D. E., 1988, Multivariate geostatistics for environmental monitoring: Sciences de la Terre. v. 27, p. 411–428.
Xie, T., 1994, Positive definite matrix-valued functions and variogram modeling: unpubl. doctoral dissertation, Univ. Arizona, 144 p.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Xie, T., Myers, D.E. Fitting matrix-valued variogram models by simultaneous diagonalization (Part I: Theory). Math Geol 27, 867–875 (1995). https://doi.org/10.1007/BF02087100
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02087100