A Generic Approach to Covariance Function Estimation Using ARMA-Models

Covariance function modeling is an essential part of stochastic methodology. Many processes in geodetic applications have rather complex, often oscillating covariance functions, where it is difficult to find corresponding analytical functions for modeling. This paper aims to give the methodological foundations for an advanced covariance modeling and elaborates a set of generic base functions which can be used for flexible covariance modeling. In particular, we provide a straightforward procedure and guidelines for a generic approach to the fitting of oscillating covariance functions to an empirical sequence of covariances. The underlying methodology is developed based on the well known properties of autoregressive processes in time series. The surprising simplicity of the proposed covariance model is that it corresponds to a finite sum of covariance functions of second-order Gauss–Markov (SOGM) processes. Furthermore, the great benefit is that the method is automated to a great extent and directly results in the appropriate model. A manual decision for a set of components is not required. Notably, the numerical method can be easily extended to ARMA-processes, which results in the same linear system of equations. Although the underlying mathematical methodology is extensively complex, the results can be obtained from a simple and straightforward numerical method.

[1]  S. Cohn,et al.  Ooce Note Series on Global Modeling and Data Assimilation Construction of Correlation Functions in Two and Three Dimensions and Convolution Covariance Functions , 2022 .

[2]  Torsten Mayer-Gürr,et al.  EGM_TIM_RL05: An independent geoid with centimeter accuracy purely based on the GOCE mission , 2014 .

[3]  S. Wu,et al.  Modeling of Continuous Stochastic Processes from Discrete Observations with Application to Sunspots Data , 1974 .

[4]  G. Sona,et al.  New Covariance Models for Local Applications of Collocation , 2001 .

[5]  A. Kolmogoroff Grundbegriffe der Wahrscheinlichkeitsrechnung , 1933 .

[6]  Helgi Tómasson Some computational aspects of Gaussian CARMA modelling , 2015, Stat. Comput..

[8]  Wolf-Dieter Schuh,et al.  Identification of Suspicious Data for Robust Estimation of Stochastic Processes , 2019 .

[9]  Alan C. Elliott,et al.  Examining Factors to Better Understand Autoregressive Models , 2009 .

[10]  T. Gneiting Correlation functions for atmospheric data analysis , 1999 .

[11]  Neda Darbeheshti,et al.  Non-stationary covariance function modelling in 2D least-squares collocation , 2009 .

[12]  Wolf-Dieter Schuh Signalverarbeitung in der Physikalischen Geodäsie , 2017 .

[13]  Fernando Sansò,et al.  Finite covariance functions , 1987 .

[14]  F. Sansò,et al.  First GOCE gravity field models derived by three different approaches , 2011 .

[15]  Steffen Schön,et al.  On the Mátern covariance family: a proposal for modeling temporal correlations based on turbulence theory , 2014, Journal of Geodesy.

[16]  P. Brockwell Continuous-time ARMA processes , 2001 .

[17]  W. Schuh,et al.  Adjustment of Digital Filters for Decorrelation of GOCE SGG Data , 2014 .

[18]  Noel A. C. Cressie,et al.  Statistics for Spatial Data: Cressie/Statistics , 1993 .

[19]  H. J. Thiébaux Anisotropic Correlation Functions for Objective Analysis , 1976 .

[20]  Estimation of the subdiurnal UT1-UTC variations by the least squares collocation method , 2000 .

[21]  Richard H. Jones FITTING A CONTINUOUS TIME AUTOREGRESSION TO DISCRETE DATA , 1981 .

[22]  Robert J. Fitzgerald Filtering Horizon-Sensor Measurements for Orbital Navigation , 1967 .

[23]  G. Tunnicliffe Wilson,et al.  Some efficient computational procedures for high order ARMA models , 1979 .

[24]  P. Broersen,et al.  How to handle colored observation noise in large least-squares problems , 2003 .

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

[26]  W. Jarmołowski,et al.  Precise estimation of covariance parameters in least-squares collocation by restricted maximum likelihood , 2014, Studia Geophysica et Geodaetica.

[27]  F. Sansò,et al.  The Analysis of Time Series with Applications to Geodetic Control Problems , 1985 .

[28]  T. Gneiting Compactly Supported Correlation Functions , 2002 .

[29]  Brandon C. Kelly,et al.  FLEXIBLE AND SCALABLE METHODS FOR QUANTIFYING STOCHASTIC VARIABILITY IN THE ERA OF MASSIVE TIME-DOMAIN ASTRONOMICAL DATA SETS , 2014, 1402.5978.

[30]  Aldo V. Vecchia,et al.  Fitting Continuous ARMA Models to Unequally Spaced Spatial Data , 1993 .

[31]  B. Friedlander,et al.  The Modified Yule-Walker Method of ARMA Spectral Estimation , 1984, IEEE Transactions on Aerospace and Electronic Systems.

[32]  S.M. Kay,et al.  Spectrum analysis—A modern perspective , 1981, Proceedings of the IEEE.

[33]  C.C.J.M. Tiberius,et al.  Noise Characteristics in High Precision GPS Positioning , 2008 .

[34]  C. Tscherning,et al.  Globally covering a-priori regional gravity covariance models , 2003 .

[35]  K. Koch,et al.  Approximating covariance matrices estimated in multivariate models by estimated auto- and cross-covariances , 2010 .

[36]  R. Forsberg,et al.  On the a priori estimation of collocation error covariance functions: a feasibility study , 2007 .

[37]  The theory of general kriging, with applications to the determination of a local geoid , 2005 .

[38]  Roland Klees,et al.  Impact of accounting for coloured noise in radar altimetry data on a regional quasi-geoid model , 2016, Journal of Geodesy.

[39]  Beate Klinger,et al.  ITSG‐Grace2018: Overview and Evaluation of a New GRACE‐Only Gravity Field Time Series , 2019, Journal of Geophysical Research: Solid Earth.