Resampling and model validation for spatio-temporal data
A03 employs complexity-reducing structural assumptions to construct resampling-based inference procedures and develops statistical tools to quantify deviations from such assumptions. A particular focus is on methods that allow to differentiate practically relevant and non-relevant deviations from simplifying assumptions such as, e.g., separability or low-rank approximations. In the long term, this project investigates resampling and model validation for multi-resolution approximations, high-dimensional modeling, and statistical inference for complex non-stationary spatio-temporal data.
Project Leaders
Prof. Dr. Holger Dette
Faculty of Mathematics - Chair of Stochastics
Ruhr University Bochum
Prof. Dr. Carsten Jentsch
Department of Statistics - Chair of Business and Social Statistics
TU Dortmund University
Summary
In spatio-temporal data analysis structural assumptions on the covariance function, such as symmetry, (spatial) isotropy or various forms of separability are often imposed to improve estimation efficiency and to achieve computational benefits, such as reduced computation time and lower data storage requirements. They are rarely made because one believes these to hold exactly, but with the hope that the deviations from the postulated model are relatively small, such that the application of more efficient statistical inferential procedures tailored for the structural assumptions is possible. Therefore, in practice, one is generally confronted with the tradeoff of balancing a potentially larger bias because of overly restrictive model assumptions against a potentially smaller variance due to the additional structure imposed.
In this project, we make use of complexity-reducing structural model assumptions to develop new and more efficient resampling-based inference procedures for the analysis of spatio-temporal data. We will develop statistical methods to quantify deviations from such assumptions and construct suitable tests for model validation. In addition to methodology for exact hypotheses (e.g. exact separability of the covariance), we will also supply methodology that allows to quantify deviations and to differentiate between practically relevant and non-relevant deviations from the postulated model assumptions. In particular, we investigate the statistical properties of these methods, such as validity or consistency, from an asymptotic point of view considering different sampling schemes in space and time.
Besides commonly used structural assumptions imposed on the covariance function such as parametric models, symmetry, (spatial) isotropy, or various forms of separability, our focus is also on complexity-reducing approaches including sparsity, low rank methods, separable expansions for covariance estimation and also nonlinear methods.
Appel, M. and E. Pebesma (2020). Spatiotemporal multi-resolution approximations for analyzing global environmental data. Spatial Statistics 38, 100465. doi: 10.1016/j.spasta.2020.100465.
Aston, J. A. D., D. Pigoli, and S. Tavakoli (2017). Tests for separability in nonparametric covariance operators of random surfaces. The Annals of Statistics 45. doi: 10.1214/16-AOS1495.
Bagchi, P. and H. Dette (2019). A Test for Separability in Covariance Operators of Random Surfaces. doi: 10.1214/19-aos1888.
Bandyopadhyay, S. and S. N. Lahiri (2009). Asymptotic properties of discrete Fourier transforms for spatial data. Sankhya: The Indian Journal of Statistics, Series A 71, 221–259. url: http://www.jstor.org/stable/41336804.
Bandyopadhyay, S., C. Jentsch, and S. Subba Rao (2017). A spectral domain test for stationarity of spatio-temporal data. Journal of Time Series Analysis 38, 326–351. doi: 10.1111/jtsa.12222.
Bandyopadhyay, S., S. N. Lahiri, and D. J. Nordman (2015). A frequency domain empirical likelihood method for irregularly spaced spatial data. The Annals of Statistics 43, 519–545. doi: 10.1214/14-AOS1291.
Bandyopadhyay, S. and S. S. Rao (2017). A Test for Stationarity for Irregularly Spaced Spatial Data. Journal of the Royal Statistical Society Series B: Statistical Methodology 79, 95–123. doi: 10.1111/rssb.12161.
Berger, J. O. and M. Delampady (1987). Testing Precise Hypotheses. Statistical Science 2, 317–335. doi: 10.1214/ss/1177013238.
Berkes, I., L. Horváth, and G. Rice (2013). Weak invariance principles for sums of dependent random functions. Stochastic Processes and their Applications 123, 385–403. doi: 10.1016/j.spa.2012.10.003.
Bucchia, B. and M. Wendler (2017). Change-point detection and bootstrap for Hilbert space valued random fields. Journal of Multivariate Analysis 155, 344–368. doi: 10.1016/j.jmva.2017.01.007.
Bücher, A., H. Dette, and F. Heinrichs (2021). Are deviations in a gradually varying mean relevant? A testing approach based on sup-norm estimators. The Annals of Statistics 49, 3583–3617. doi: 10.1214/21-AOS2098.
Castillo-P\´aez, S., R. Fern\´andez-Casal, and P. García-Soid\´an (2019). A nonparametric bootstrap method for spatial data. Computational Statistics & Data Analysis 137, 1–15. doi: 10.1016/j.csda.2019.01.017.
Charkaborty, A. and V. M. Panaretos (2022). Testing for the rank of a covariance operator. The Annals of Statistics 50, 3510–3537. doi: 10.1214/22-AOS2238.
Chen, R., H. Xiao, and D. Yang (2021a). Autoregressive models for matrix-valued time series. Journal of Econometrics 222, 539–560. doi: 10.1016/j.jeconom.2020.07.015.
Chen, W., M. Genton, and Y. Sun (2021b). Space-time covariance structures and models. Annual Review of Statistics and Its Application 8, 191–215. doi: 10.1146/annurev-statistics-042720-115603.
Constantinou, P. Kokoszka, and . Reimherr (2017). Testing separability of space-time functional processes. Biometrika. doi: 10.1093/biomet/asx013.
Cressie, N. and H.-C. Huang (1999). Classes of nonseparable, spatio-temporal stationary covariance functions. Journal of the American Statistical Association 94, 1330–1339. doi: 10.1080/01621459.1999.10473885.
Cressie, N. and G. Johannesson (2008). Fixed rank kriging for very large spatial data sets. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 70, 209–226. doi: 10.1111/j.1467-9868.2007.00633.x.
Crujeiras, R. M., R. Fernández-Casal, and W. González-Manteiga (2010). Nonparametric test for separability of spatio-temporal processes. Environmetrics 21, 382–399. doi: 10.1002/env.1006.
Dette, H., G. Dierickx, and T. Kutta (2022). Quantifying deviations from separability in space-time functional processes. Bernoulli 28, 2909–2940. doi: 10.3150/21-BEJ1442.
Dette, H. and K. Kokot (2020). Bio-equivalence tests in functional data by maximum deviation. Biometrika 108, 895–913. doi: 10.1093/biomet/asaa096.
Dette, H. and K. Kokot (2022). Detecting relevant differences in the covariance operators of functional time series: A sup-norm approach. Annals of the Institute of Statistical Mathematics 74, 195–231. doi: 10.1007/s10463-021-00795-2.
Dette, H., K. Kokot, and A. Aue (2020a). Functional data analysis in the Banach space of continuous functions. The Annals of Statistics 48. doi: 10.1214/19-AOS1842.
Dette, H., K. Kokot, and S. Volgushev (2020b). Testing Relevant Hypotheses in Functional Time Series via Self-Normalization. Journal of the Royal Statistical Society Series B: Statistical Methodology 82, 629–660. doi: 10.1111/rssb.12370.
Dette, H., K. Möllenhoff, S. Volgushev, and F. Bretz (2018). Equivalence of regression curves. Journal of the American Statistical Association 113, 711–729. doi: 10.1080/01621459.2017.1281813.
Dette, H. and P. Quanz (2023). Detecting relevant changes in the spatiotemporal mean function. Journal of Time Series Analysis. doi: 10.1111/jtsa.12674.
Dette, H. and D. Wied (2015). Detecting relevant changes in time series models. Journal of the Royal Statistical Society Series B: Statistical Methodology 78, 371–394. doi: 10.1111/rssb.12121.
Dette, H. and W. Wu (2019). Detecting relevant changes in the mean of nonstationary processes‚ a mass excess approach. The Annals of Statistics 47, 3578–3608. doi: 10.1214/19-AOS1811.
Fogarty, C. B. and D. S. Small (2014). Equivalence testing for functional data with an application to comparing pulmonary function devices. The Annals of Applied Statistics 8, 2002–2026. doi: 10.1214/14-AOAS763.
Fuentes, M. (2006). Testing for separability of spatial-temporal covariance functions. Journal of Statistical Planning and Inference 136, 447–466. doi: 10.1016/j.jspi.2004.07.004.
Genton, M. G. (2007). Separable approximations of space-time covariance matrices. Environmetrics 18, 681–695. doi: 10.1002/env.854.
Gneiting, T. (2002). Nonseparable, Stationary Covariance Functions for Space–Time Data. Journal of the American Statistical Association 97, 590–600. doi: 10.1198/016214502760047113.
Gneiting, T., M. Genton, and P. Guttorp (2006). Geostatistical space-time models, stationarity, separability, and full symmetry. Monographs on Statistics and Applied Probability 107, 151–175. doi: 10.1201/9781420011050.
Greenewald, K. and A. O. Hero (2015). Robust Kronecker product PCA for spatio-temporal covariance estimation. IEEE Transactions on Signal Processing 63, 6368–6378. doi: 10.1109/TSP.2015.2472364.
Guan, Y., M. Sherman, and J. A. Calvin (2004). A nonparametric test for spatial isotropy using subsampling. Journal of the American Statistical Association 99, 810–821. doi: 10.1198/016214504000001150.
Hafner, C. M., O. B. Linton, and H. Tang (2020). Estimation of a multiplicative correlation structure in the large dimensional case. Journal of Econometrics 217, 431–470. doi: 10.1016/j.jeconom.2019.12.012.
Hala, M. van, S. Bandyopadhyay, S. N. Lahiri, and D. J. Nordman (2020). A general frequency domain method for assessing spatial covariance structures. Bernoulli 26. doi: 10.3150/19-BEJ1160.
Heaton, M. J., A. Datta, A. O. Finley, R. Furrer, et al. (2019). A Case Study Competition Among Methods for Analyzing Large Spatial Data. Journal of Agricultural, Biological and Environmental Statistics 24, 398–425. doi: 10.1007/s13253-018-00348-w.
Heinrichs, F. and H. Dette (2021). A distribution free test for changes in the trend function of locally stationary processes. Electronic Journal of Statistics, 3762–3797. doi: 10.1214/21-EJS1871.
Hörmann, S., L. Kidzinski, and M. Hallin (2014). Dynamic functional principal components. Journal of the Royal Statistical Society Series B: Statistical Methodology 77, 319–348. doi: 10.1111/rssb.12076.
Hsu, N.-J., H.-C. Huang, and R. S. Tsay (2021). Matrix autoregressive spatio-temporal models. Journal of Computational and Graphical Statistics 30, 1143–1155. doi: 10.1080/10618600.2021.1938587.
Jentsch, C. and M. Meyer (2021). On the validity of Akaike’s identity for random fields. Journal of Econometrics 222, 676–687. doi: 10.1016/j.jeconom.2020.04.044.
Jentsch, C. and D. N. Politis (2015). Covariance matrix estimation and linear process bootstrap for multivariate time series of possibly increasing dimension. The Annals of Statistics 43. doi: 10.1214/14-AOS1301.
Jentsch, C. and S. Subba Rao (2015). A test for second order stationarity of a multivariate time series. Journal of Econometrics 185, 124–161. doi: 10.1016/j.jeconom.2014.09.010.
Jun, M. and M. G. Genton (2012). A test for stationarity of spatio-temporal random fields on planar and spherical domains. Statistica Sinica 22, 1737–1764. doi: 10.5705/ss.2010.251.
Katzfuss, M. (2017). A multi-resolution approximation for massive spatial datasets. Journal of the American Statistical Association 112, 201–214. doi: 10.1080/01621459.2015.1123632.
Kokoszka, P. and M. Reimherr (2013). Asymptotic normality of the principal components of functional time series. Stochastic Processes and their Applications 123, 1546–1562. doi: 10.1016/j.spa.2012.12.011.
Krampe, J., J.-P. Kreiss, and E. Paparoditis (2017). Estimated Wold representation and spectral density-driven bootstrap for time series. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 80, 703–726. doi: 10.1111/rssb.12267.
Krampe, J. and T. L. McMurry (2021). Estimating wold matrices and vector moving average processes. Journal of Time Series Analysis 42, 201–221. doi: 10.1111/jtsa.12562.
Kurisu, D., K. Kato, and X. Shao (2023). Gaussian approximation and spatially dependent wild bootstrap for high-dimensional spatial data. Journal of the American Statistical Association, 1–21. doi: 10.1080/01621459.2023.2218578.
Lahiri, S. N. (2003). Central limit theorems for weighted sums of a spatial process under a Class of stochastic and fixed designs. Sankhya: The Indian Journal of Statistics (2003-2007) 65, 356–388. url: http://www.jstor.org/stable/25053269.
Li, B., M. G. Genton, and M. Sherman (2007). A nonparametric assessment of properties of space-time covariance functions. Journal of the American Statistical Association 102, 736–744. doi: 10.1198/016214507000000202.
Liang, F., Y. Cheng, Q. Song, J. Park, et al. (2013). A Resampling-Based Stochastic Approximation Method for Analysis of Large Geostatistical Data. Journal of the American Statistical Association 108, 325–339. doi: 10.1080/01621459.2012.746061.
Lindgren, F., H. Rue, and J. Lindström (2011). An explicit link between Gaussian fields and Gaussian Markov random fields: The stochastic partial differential equation approach. Journal of the Royal Statistical Society Series B 73, 423–498. doi: 10.1111/j.1467-9868.2011.00777.x.
Linton, O. B. and H. Tang (2022). ESTIMATION OF THE KRONECKER COVARIANCE MODEL BY QUADRATIC FORM. Econometric Theory 38, 1014–1067. doi: 10.1017/S026646662000050X.
Liu, S. and G. Trenkler (2008). Hadamard, Khatri-Rao, Kronecker and other matrix products. International Journal of Information and Systems Sciences 4, 160–177.
Lobato, I. N. (2001). Testing that a dependent process is uncorrelated. Journal of the American Statistical Association 96, 1066–1076. doi: 10.1198/016214501753208726.
Loh, J. M. and M. L. Stein (2008). Spatial bootstrap with increasing observations in a fixed domain. Statistica Sinica 18, 667–688.
Lu, N. and D. L. Zimmerman (2005). Testing for directional symmetry in spatial dependence using the periodogram. Journal of Statistical Planning and Inference 129, 369–385. doi: 10.1016/j.jspi.2004.06.058.
Lynch, B. and K. Chen (2018). A test of weak separability for multi-way functional data, with application to brain connectivity studies. Biometrika 105, 815–831. doi: 10.1093/biomet/asy048.
Ma, C. (2003). Spatio-temporal stationary covariance models. Journal of Multivariate Analysis 86, 97–107. doi: 10.1016/S0047-259X(02)00014-3.
Maity, A. and M. Sherman (2012). Testing for spatial isotropy under general designs. Journal of Statistical Planning and Inference 142, 1081–1091. doi: 10.1016/j.jspi.2011.11.013.
Masak, T., S. Sarkar, and V. Panaretos (2022). Separable expansions for covariance estimation via the partial inner product. Biometrika 110, 225–247. doi: 10.1093/biomet/asac035.
Matsuda, Y. and Y. Yajima (2004). On testing for separable correlations of multivariate time series. Journal of Time Series Analysis 25, 501–528. doi: 10.1111/j.1467-9892.2004.01795.x.
Matsuda, Y. and Y. Yajima (2009). Fourier analysis of irregularly spaced data on \mathbbR^d. Journal of the Royal Statistical Society Series B 71, 191–217. doi: 10.1111/j.1467-9868.2008.00685.x.
McMurry, T. L. and D. N. Politis (2010). Banded and tapered estimates for autocovariance matrices and the linear process bootstrap. Journal of Time Series Analysis 31, 471–482. doi: 10.1111/j.1467-9892.2010.00679.x.
Meyer, M., C. Jentsch, and J.-P. Kreiss (2017). Baxter’s inequality and sieve bootstrap for random fields. Bernoulli 23. doi: 10.3150/16-BEJ835.
Mitchell, M. W., M. G. Genton, and M. L. Gumpertz (2005). Testing for separability of space–time covariances. Environmetrics 16, 819–831. doi: 10.1002/env.737.
Mitchell, M. W., M. G. Genton, and M. L. Gumpertz (2006). A likelihood ratio test for separability of covariances. Journal of Multivariate Analysis 97, 1025–1043. doi: 10.1016/j.jmva.2005.07.005.
Ng, W. L., C. Y. Yau, and X. Chen (2021). Frequency domain bootstrap methods for random fields. Electronic Journal of Statistics 15, 6586–6632. doi: 10.1214/21-EJS1959.
Nordman, D. J. and S. N. Lahiri (2004). On optimal spatial subsample size for variance estimation. The Annals of Statistics 32, 1981–2027. doi: 10.1214/009053604000000779.
Nordman, D. J., S. N. Lahiri, and B. L. Fridley (2007). Optimal block size for variance estimation by a spatial block bootstrap method. Sankhya: The Indian Journal of Statistics 69, 468–493. url: http://www.jstor.org/stable/25664573.
Politis, D. and J. Romano (1993). Nonparametric Resampling for Homogeneous Strong Mixing Random Fields. Journal of Multivariate Analysis 47, 301–328. doi: 10.1006/jmva.1993.1085.
Politis, D. N., E. Paparoditis, and J. P. Romano (1998). Large sample inference for irregularlyspaced dependent observations based on subsampling. Sankya Series A 60, 274–292.
Politis, D. N., J. P. Romano, and M. Wolf (1999). Subsampling. New York, Springer. doi: 10.1007/978-1-4612-1554-7.
Porcu, E., R. Furrer, and D. Nychka (2021). 30 Years of space-time covariance functions. WIREs Computational Statistics 13. doi: 10.1002/wics.1512.
Rabinowicz, A. and S. Rosset (2021). Resampling methods for detecting anisotropic correlation structure.
Reichold, K. and C. Jentsch (2022). A Bootstrap-Assisted Self-Normalization Approach to Inference in Cointegrating Regressions. url: http://arxiv.org/abs/2204.01373.
Reichold, K. and C. Jentsch (2023). Bootstrap inference in cointegrating regressions: Traditional and self-normalized test statistics. Journal of Business & Economic Statistics, 1–97. doi: 10.1080/07350015.2023.2271538.
Saha, A. and A. Datta (2018). BRISC: Bootstrap for rapid inference on spatial covariances. Stat 7. doi: 10.1002/sta4.184.
Scaccia, L. and R. J. Martin (2005). Testing axial symmetry and separability of lattice processes. Journal of Statistical Planning and Inference 131, 19–39. doi: 10.1016/j.jspi.2004.01.003.
Shao, X. (2010). A self-normalized approach to confidence interval construction in time series. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 72, 343–366. doi: 10.1111/j.1467-9868.2009.00737.x.
Shao, X. (2015). Self-Normalization for Time Series: A Review of Recent Developments. Journal of the American Statistical Association 110, 1797–1817. doi: 10.1080/01621459.2015.1050493.
Shao, X. and B. Li (2009). A tuning parameter free test for properties of space–time covariance functions. Journal of Statistical Planning and Inference 139, 4031–4038. doi: 10.1016/j.jspi.2009.05.008.
Sherman, M. (2011). Spatial Statistics and Spatio-Temporal Data: Covariance Functions and Directional Properties. Chichester, West Sussex, U.K: Wiley. doi: 10.1002/9780470974391.
Song, D. and A. O. Hero (2023). On Kronecker separability of multiway covariance.
Stein, M. L. (2005). Space–Time Covariance Functions. Journal of the American Statistical Association 100, 310–321. doi: 10.1198/016214504000000854.
Sun, H. and Y. Chen (2023). Matrix autoregressive model with vector time series covariates for spatio-temporal data.
Tang, Y., H. J. Wang, Y. Sun, and A. S. Hering (2019). Copula-based semiparametric models for spatiotemporal data. Biometrics 75, 1156–1167. doi: 10.1111/biom.13066.
Tukey, J. W. (1991). The philosophy of multiple comparisons. Statistical Science 6, 100–116. doi: 10.1214/ss/1177011945.
Van Loan, C. F. and N. Pitsianis (1993). Approximation with Kronecker Products. Linear Algebra for Large Scale and Real-Time Applications. Dordrecht: Springer Netherlands, 293–314. doi: 10.1007/978-94-015-8196-7_17.
Wellek, S. (2010). Testing Statistical Hypotheses of Equivalence and Noninferiority. CRC Press. doi: 10.1201/EBK1439808184.
Weller, Z. D. and J. A. Hoeting (2020). A nonparametric spectral domain test of spatial isotropy. Journal of Statistical Planning and Inference 204, 177–186. doi: 10.1016/j.jspi.2019.05.005.
Zhang, X. and X. Shao (2015). Two sample inference for the second-order property of temporally dependent functional data. Bernoulli 21, 909–929. doi: 10.3150/13-BEJ592.