Towards an automatic procedure for modeling multivariate space-time data

In many environmental sciences, several correlated variables are observed at some locations of the domain of interest and over a certain period of time. In this context, appropriate modeling and prediction techniques for multivariate space-time data as well as interactive software packages are necessary. In this paper, a new automatic procedure for fitting the space-time linear coregionalization model (ST-LCM) using the product-sum variogram model is discussed. This procedure, based on the simultaneous diagonalization of the sample matrix variograms, allows the identification of the ST-LCM parameters in a very flexible way. The fitting process is analytically described by a main flow chart and all steps are specified by four subprocedures. An application of this procedure is illustrated through a case study concerning the daily concentrations of three air pollutants measured in an urban area. Then the fitted space-time coregionalization model is applied to predict the variable of interest using a recent GSLib routine, named ''COK2ST.''

[1]  Clayton V. Deutsch,et al.  An intrinsic model of coregionalization that solves variance inflation in collocated cokriging , 2009, Comput. Geosci..

[2]  Sandra De Iaco,et al.  Space–time correlation analysis: a comparative study , 2010 .

[3]  R. M. Lark,et al.  Fitting a linear model of coregionalization for soil properties using simulated annealing , 2003 .

[4]  Tatiyana V. Apanasovich,et al.  Cross-covariance functions for multivariate random fields based on latent dimensions , 2010 .

[5]  B. Pelletier,et al.  Fitting the Linear Model of Coregionalization by Generalized Least Squares , 2004 .

[6]  Chunsheng Ma,et al.  Spatio-Temporal Covariance Functions Generated by Mixtures , 2002 .

[7]  Donato Posa,et al.  Using Simultaneous Diagonalization to Identify a Space–Time Linear Coregionalization Model , 2012, Mathematical Geosciences.

[8]  S. De Iaco,et al.  Modeling and prediction of multivariate space-time random fields , 2005, Comput. Stat. Data Anal..

[9]  T. Gneiting.  Nonseparable, Stationary Covariance Functions for Space–Time Data , 2002 .

[10]  Timothy C. Coburn,et al.  Geostatistics for Natural Resources Evaluation , 2000, Technometrics.

[11]  T. Gneiting,et al.  Matérn Cross-Covariance Functions for Multivariate Random Fields , 2010 .

[12]  D. E. Myers,et al.  Strict Positive Definiteness of a Product of Covariance Functions , 2011 .

[13]  Donato Posa,et al.  FORTRAN programs for space-time modeling , 2002 .

[14]  D. Myers,et al.  Space–time analysis using a general product–sum model , 2001 .

[15]  S. De Iaco,et al.  Space-time variograms and a functional form for total air pollution measurements , 2002, Comput. Stat. Data Anal..

[16]  Xavier Emery,et al.  Iterative algorithms for fitting a linear model of coregionalization , 2010, Comput. Geosci..

[17]  John Walden,et al.  The use of simultaneous R- and Q-mode factor analysis as a tool for assisting interpretation of mineral magnetic data , 1992 .

[18]  N. Cressie,et al.  Classes of nonseparable, spatio-temporal stationary covariance functions , 1999 .

[19]  M. Goulard,et al.  Linear coregionalization model: Tools for estimation and choice of cross-variogram matrix , 1992 .

[20]  S. D. Iaco,et al.  A new space--time multivariate approach for environmental data analysis , 2011 .

[21]  Clayton V. Deutsch,et al.  GSLIB: Geostatistical Software Library and User's Guide , 1993 .

[22]  Jerry M. Davis,et al.  Multivariate Spatial-temporal Modeling and Prediction of Speciated Fine Particles , 2009, Journal of statistical theory and practice.

[23]  Shahrokh Rouhani,et al.  Multivariate geostatistical approach to space‐time data analysis , 1990 .

[24]  Chunsheng Ma,et al.  Recent developments on the construction of spatio-temporal covariance models , 2008 .

[25]  Antoine Souloumiac,et al.  Jacobi Angles for Simultaneous Diagonalization , 1996, SIAM J. Matrix Anal. Appl..

[26]  P. Goovaerts,et al.  Study of Spatial and Temporal Variations of Hydrogeochemical Variables Using Factorial Kriging Analysis , 1993 .

[27]  Bo Li,et al.  Testing the covariance structure of multivariate random fields , 2008 .

[28]  Dionissios T. Hristopulos,et al.  Methods for generating non-separable spatiotemporal covariance models with potential environmental applications , 2004 .

[29]  M. Stein.  Space–Time Covariance Functions , 2005 .

[30]  S. De Iaco,et al.  FORTRAN programs for space-time multivariate modeling and prediction , 2010, Comput. Geosci..

[31]  D. E. Myers,et al.  The Linear Coregionalization Model and the Product–Sum Space–Time Variogram , 2003 .

[32]  D. Myers,et al.  Fitting matrix-valued variogram models by simultaneous diagonalization (Part I: Theory) , 1995 .

[33]  Alan E. Gelfand,et al.  Spatial process modelling for univariate and multivariate dynamic spatial data , 2005 .

[34]  Donato Posa,et al.  On strict positive definiteness of product and productsum covariance models , 2011 .

[35]  Milt Statheropoulos,et al.  Principal component and canonical correlation analysis for examining air pollution and meteorological data , 1998 .

[36]  Jorge Mateu,et al.  New classes of covariance and spectral density functions for spatio-temporal modelling , 2008 .

[37]  Hao Zhang,et al.  Maximum‐likelihood estimation for multivariate spatial linear coregionalization models , 2007 .