Estimating and modeling space-time correlation structures

被引：116

作者：

De Cesare, L

Myers, DE

Posa, D

机构：

[1] Univ Arizona, Dept Math, Tucson, AZ 85721 USA

[2] Univ Bari, Fac Econ 2, I-70121 Bari, Italy

[3] CNR, Ist Ric Matemat Applicata, I-70126 Bari, Italy

[4] Univ Lecce, Fac Econ, I-73100 Lecce, Italy

来源：

STATISTICS & PROBABILITY LETTERS | 2001年 / 51卷 / 01期

关键词：

space-time correlation; geostatistics;

D O I：

10.1016/S0167-7152(00)00131-0

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

In this paper, a class of product-sum covariance models has been introduced for estimating and modeling space-time correlation structures. It is shown how the coefficients of this class of models are related to the global sill. and "partial" spatial and temporal sills; moreover, some constraints on these sills have been given in order to assure positive definiteness of the product-sum covariance model. A brief comparative study with some other classes of spatial-temporal covariance models has been pointed out. (C) 2001 Elsevier Science B.V. All rights reserved.

引用

页码：9 / 14

页数：6

共 17 条

[1]

[Anonymous], 1981, MINING GEOSTATISTICS

[2]

[Anonymous], GEOSTATISTICS

[3]

[Anonymous], GEOSTATISTICS WOLLON

[4]

[Anonymous], CAN J STAT, DOI [10.2307/3314707, DOI 10.2307/3314707]