Discussing the “big n problem”
- 465 Downloads
When a large amount of spatial data is available computational and modeling challenges arise and they are often labeled as “big n problem”. In this work we present a brief review of the literature. Then we focus on two approaches, respectively based on stochastic partial differential equations and integrated nested Laplace approximation, and on the tapering of the spatial covariance matrix. The fitting and predictive abilities of using the two methods in conjunction with Kriging interpolation are compared in a simulation study.
KeywordsSPDE INLA Tapering Large spatial data sets Spatial statistics
Unable to display preview. Download preview PDF.
- Banerjee S, Carlin BP, Gelfand AE (2004) Hierarchical modeling and analysis for spatial data. Chapman and Hall, LondonGoogle Scholar
- Bolin D, Lindgren F (2011) Spatial wavelet Markov models are more efficient than covariance tapering and process convolutions. arXiv:11061980v1. http://arxiv.org/abs/1106.1980,1106.1980
- Brenner SC, Scott R: The mathematical theory of finite element methods. Springer, Berlin (2007)Google Scholar
- Higdon D, Swall J, Kern J: Non-stationary spatial modeling. Bayesian Stat 6, 761–768 (1998)Google Scholar
- Ji WY, Simon W, Koray K, Ercan EK: Variant functional approximations for latent Gaussian models. Technical Report of Statistics Department. Trinity College, Dublin (2011)Google Scholar
- Rue H, Tjelmeland H (2002) Fitting Gaussian Markov random fields to Gaussian fields. Scand J Stat, pp 31–49. doi: 10.1111/1467-9469.00058
- Shaby B, Ruppert D (2012) Tapered covariance: Bayesian estimation and asymptotics. J Comput Graph Stat 21:433–452Google Scholar
- Whittle P (1954) On stationary processes in the plane. Biometrika 41. doi: 10.2307/2332724
- Zhang H, Du J: Covariance tapering in spatial statistics. In: Mateu, E, Porcu, J (eds) Positive definite functions from Schoenberg to space-time challenges, Gráficas Castañ, s.l. (2008)Google Scholar