fgasp-class: Fast GaSP class

In FastGaSP: Fast and Exact Computation of Gaussian Stochastic Process

Fast GaSP class

Description

S4 class for fast computation of the Gaussian stochastic process (GaSP) model with the Matern kernel function with or without a noise.

Objects from the Class

Objects of this class are created and initialized with the function fgasp that computes the calculations needed for setting up the estimation and prediction.

Slots

num_obs:

object of class integer. The number of experimental observations.

have_noise:

object of class logical to specify whether the the model has a noise or not. "TRUE" means the model contains a noise and "FALSE" means the model does not contain a noise.

kernel_type:

a character to specify the type of kernel to use.The current version supports kernel_type to be "matern_5_2" or "exp", meaning that the matern kernel with roughness parameter being 2.5 or 0.5 (exponent kernel), respectively.

input:

object of class vector with dimension num_obs x 1 for the sorted input locations.

delta_x:

object of class vector with dimension (num_obs-1) x 1 for the differences between the sorted input locations.

output:

object of class vector with dimension num_obs x 1 for the observations at the sorted input locations.

Methods

show

Prints the main slots of the object.

predict

See predict.

Author(s)

Mengyang Gu [aut, cre], Xinyi Fang [aut], Yizi Lin [aut]

Maintainer: Mengyang Gu <[e-mail]>

References

Hartikainen, J. and Sarkka, S. (2010). Kalman filtering and smoothing solutions to temporal gaussian process regression models, Machine Learning for Signal Processing (MLSP), 2010 IEEE International Workshop, 379-384.

M. Gu, Y. Xu (2017), Nonseparable Gaussian stochastic process: a unified view and computational strategy, arXiv:1711.11501.

M. Gu, X. Wang and J.O. Berger (2018), Robust Gaussian Stochastic Process Emulation, Annals of Statistics, 46, 3038-3066.

See Also

fgasp for more details about how to create a fgasp object.

FastGaSP documentation built on April 4, 2025, 5:16 a.m.