×

Approximate inference for spatial functional data on massively parallel processors. (English) Zbl 1506.62152

Summary: With continually increasing data sizes, the relevance of the big \(n\) problem of classical likelihood approaches is greater than ever. The functional mixed-effects model is a well established class of models for analyzing functional data. Spatial functional data in a mixed-effects setting is considered, and so-called operator approximations for doing inference in the resulting models are presented. These approximations embed observations in function space, transferring likelihood calculations to the functional domain. The resulting approximated problems are naturally parallel and can be solved in linear time. An extremely efficient GPU implementation is presented, and the proposed methods are illustrated by conducting a classical statistical analysis of 2D chromatography data consisting of more than 140 million spatially correlated observation points.

MSC:

62-08 Computational methods for problems pertaining to statistics
62R10 Functional data analysis

Software:

cudaBayesreg; gss; R; fda (R)
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Boularan, J.; Ferré, L.; Vieu, P., Growth curves: a two-stage nonparametric approach, Journal of Statistical Planning and Inference, 38, 3, 327-350, (1994) · Zbl 0797.62030
[2] Chen, H.; Wang, Y., A penalized spline approach to functional mixed effects model analysis, Biometrics, 67, 3, 861-870, (2011) · Zbl 1226.62030
[3] Craven, P.; Wahba, G., Smoothing noisy data with spline functions, Numerische Mathematik, 31, 377-403, (1978) · Zbl 0377.65007
[4] da Silva, A. F., Cudabayesreg: Bayesian computation in CUDA, The R Journal, 2, 2, 48-55, (2010)
[5] Dolph, C. L.; Woodbury, M. A., On the relation between green’s functions and covariances of certain stochastic processes and its application to unbiased linear prediction, Transactions of the American Mathematical Society, 519-550, (1952) · Zbl 0048.11202
[6] Duffy, D. G., Green’s functions with applications, (2001), Chapman & Hall/CRC · Zbl 0983.35003
[7] Ferraty, F.; Vieu, P., Nonparametric functional data analysis, (2006), Springer · Zbl 1119.62046
[8] Grewenig, S.; Weickert, J.; Bruhn, A., From box filtering to fast explicit diffusion, (Goesele, M.; Roth, S.; Kuijper, A.; Schiele, B.; Schindler, K., Pattern Recognition, Lecture Notes in Computer Science, vol. 6376, (2010), Springer Berlin, Heidelberg), 533-542
[9] Guo, W., Functional mixed effects models, Biometrics, 58, 1, 121-128, (2002) · Zbl 1209.62072
[10] Hartung, S.; Shukla, H.; Miller, J. P.; Pennypacker, C., GPU acceleration of image convolution using spatially-varying kernel, (2012 19th IEEE International Conference on Image Processing, ICIP, (2012), IEEE), 1685-1688
[11] Harville, D. A., Maximum likelihood approaches to variance component estimation and to related problems, Journal of the American Statistical Association, 72, 320-340, (1977) · Zbl 0373.62040
[12] Horváth, L.; Kokoszka, P., Inference for functional data with applications, vol. 200, (2012), Springer
[13] Jordan, M. I., Message from the president: what are the open problems in Bayesian statistics?, ISBA Bulletin, 18, 1-4, (2011)
[14] Lee, D.-J.; Durbán, M.; Eilers, P., Efficient two-dimensional smoothing with \(P\)-spline ANOVA mixed models and nested bases, Computational Statistics & Data Analysis, 61, 22-37, (2013) · Zbl 1348.65030
[15] Lee, Y.; Nelder, J. A.; Pawitan, Y., Generalized linear models with random effects: unified analysis via \(H\)-likelihood, (2006), Chapman & Hall · Zbl 1110.62092
[16] Lindgren, F.; Rue, H.; Lindström, J., 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 (Statistical Methodology), 73, 4, 423-498, (2011) · Zbl 1274.62360
[17] Liu, Z.; Guo, W., Functional mixed effects models, Wiley Interdisciplinary Reviews: Computational Statistics, 4, 6, 527-534, (2012)
[18] Markussen, B., Functional data analysis in an operator-based mixed-model framework, Bernoulli, 19, 1-17, (2013) · Zbl 1259.62001
[19] Micikevicius, P., 3D finite difference computation on GPUs using CUDA, (Proceedings of 2nd Workshop on General Purpose Processing on Graphics Processing Units, (2009), ACM), 79-84
[20] Núñez-Antón, V.; Rodríguez-Póo, J. M.; Vieu, P., Longitudinal data with nonstationary errors: a nonparametric three-stage approach, Test, 8, 1, 201-231, (1999) · Zbl 0945.62042
[21] Pawitan, Y., In all likelihood, (2001), Oxford University Press · Zbl 1013.62001
[22] Petersen, I. L.; Tomasi, G.; Sørensen, H.; Boll, E. S.; Hansen, H. C.B.; Christensen, J. H., The use of environmental metabolomics to determine glyphosate level of exposure in rapeseed (brassica napus L.) seedlings, Environmental Pollution, 159, 10, 3071-3077, (2011)
[23] Pintore, A.; Speckman, P.; Holmes, C. C., Spatially adaptive smoothing splines, Biometrika, 93, 1, 113-125, (2006) · Zbl 1152.62331
[24] R: A language and environment for statistical computing, (2012), R foundation for statistical computing Vienna, Austria, URL http://www.R-project.org/
[25] Rakêt, L.L., 2013. Duality based optical flow algorithms with applications. University of Copenhagen prize thesis in Computer Science, Copenhagen University Library.
[26] Rakêt, L. L.; Roholm, L.; Nielsen, M.; Lauze, F., TV-\(L^1\) optical flow for vector valued images, (Boykov, Y.; Kahl, F.; Lempitsky, V.; Schmidt, F., Energy Minimization Methods in Computer Vision and Pattern Recognition, Lecture Notes in Computer Science, vol. 6819, (2011), Springer), 329-343
[27] Ramsay, J. O.; Silverman, B. W., Functional data analysis, (2005), Springer · Zbl 1079.62006
[28] Robinson, G. K., That BLUP is a good thing: the estimation of random effects, Statistical Science, 6, 1, 15-32, (1991) · Zbl 0955.62500
[29] Staicu, A.-M.; Crainiceanu, C. M.; Carroll, R. J., Fast methods for spatially correlated multilevel functional data, Biostatistics, 11, 2, 177-194, (2010)
[30] Suchard, M. A.; Wang, Q.; Chan, C.; Frelinger, J.; Cron, A.; West, M., Understanding GPU programming for statistical computation: studies in massively parallel massive mixtures, Journal of Computational and Graphical Statistics, 19, 2, 419-438, (2010)
[31] Wahba, G., Spline models for observational data, (1990), Society for Industrial and Applied Mathematics · Zbl 0813.62001
[32] Wahba, G.; Wendelberger, J., Some new mathematical methods for variational objective analysis using splines and cross validation, Monthly Weather Review, 108, 1122, (1980)
[33] Wang, Y., Mixed effects smoothing spline analysis of variance, Journal of the Royal Statistical Society: Series B (Statistical Methodology), 60, 1, 159-174, (1998) · Zbl 0909.62034
[34] Weickert, J.; Schnörr, C., Variational optic flow computation with a spatio-temporal smoothness constraint, Journal of Mathematical Imaging and Vision, 14, 245-255, (2001) · Zbl 0988.68821
[35] Yue, Y.R., Simpson, D., Lindgren, F., Rue, H., 2012a. Bayesian adaptive smoothing spline using stochastic differential equations. ArXiv preprint arXiv:1209.2013.
[36] Yue, Y. R.; Speckman, P. L.; Sun, D., Priors for Bayesian adaptive spline smoothing, Annals of the Institute of Statistical Mathematics, 64, 3, 577-613, (2012) · Zbl 1237.62037
[37] Zach, C.; Pock, T.; Bischof, H., A duality based approach for realtime TV-\(L^1\) optical flow, (Hamprecht, F.; Schnörr, C.; Jähne, B., Pattern Recognition, Lecture Notes in Computer Science, vol. 4713, (2007), Springer), 214-223
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.