A general framework for Vecchia approximations of Gaussian processes

@article{Katzfuss2017AGF,
  title={A general framework for Vecchia approximations of Gaussian processes},
  author={Matthias Katzfuss and J. Guinness},
  journal={arXiv: Methodology},
  year={2017}
}
  • Matthias Katzfuss, J. Guinness
  • Published 2017
  • Mathematics
  • arXiv: Methodology
  • Gaussian processes (GPs) are commonly used as models for functions, time series, and spatial fields, but they are computationally infeasible for large datasets. Focusing on the typical setting of modeling data as a GP plus an additive noise term, we propose a generalization of the Vecchia (1988) approach as a framework for GP approximations. We show that our general Vecchia approach contains many popular existing GP approximations as special cases, allowing for comparisons among the different… CONTINUE READING
    63 Citations
    Vecchia Approximations of Gaussian-Process Predictions
    • 22
    • PDF
    Vecchia-Laplace approximations of generalized Gaussian processes for big non-Gaussian spatial data
    • 8
    • PDF
    Scaled Vecchia approximation for fast computer-model emulation
    • 7
    • PDF
    Scalable Gaussian Process Computations Using Hierarchical Matrices
    • 12
    • PDF
    A class of multi-resolution approximations for large spatial datasets
    • 16
    • PDF
    Bayesian Inference for Big Spatial Data Using Non-stationary Spectral Simulation
    • 1
    • Highly Influenced
    • PDF
    Bayesian nonstationary and nonparametric covariance estimation for large spatial data.
    • PDF
    2 Infill Asymptotics for Vecchia ’ s approximation with Matérn model 2
    • Highly Influenced
    Bayesian inference for high-dimensional nonstationary Gaussian processes
    • 1
    • Highly Influenced
    • PDF

    References

    SHOWING 1-10 OF 59 REFERENCES
    Vecchia Approximations of Gaussian-Process Predictions
    • 22
    • PDF
    Local and global sparse Gaussian process approximations
    • 215
    • PDF
    On nearest-neighbor Gaussian process models for massive spatial data.
    • 31
    • Highly Influential
    Permutation and Grouping Methods for Sharpening Gaussian Process Approximations
    • J. Guinness
    • Mathematics, Computer Science
    • Technometrics
    • 2018
    • 45
    • PDF
    Statistically and Computationally Efficient Estimating Equations for Large Spatial Datasets
    • 31
    • PDF
    Smoothed Full-Scale Approximation of Gaussian Process Models for Computation of Large Spatial Datasets
    • 8
    • Highly Influential
    • PDF
    Limitations on low rank approximations for covariance matrices of spatial data
    • 127
    A class of multi-resolution approximations for large spatial datasets
    • 16
    • PDF
    Approximating likelihoods for large spatial data sets
    • 361
    • Highly Influential
    • PDF