gp_mcmc: Basic regression in Gaussian processes

In cboettig/nonparametric-bayes: Nonparametric Bayes inference for ecological models

Description

Basic regression in Gaussian processes

Usage

gp_mcmc(x, y, init_pars = c(l = 1, sigma.n = 1), n = 10000, d.p = c(5, 5),
  s2.p = c(5, 5))

Arguments

x	Observed x values, (vector or matrix with columns for each dimension of data)
y	Vector of observed y values in the training data
init_pars	the initial guesses for lengthscale l and process noise sigma_n
n	iterations of the metropolis algorithm
d.p	parameters for the length-scale prior, as an inverse Gamma distribution
s2.p	parameters for the noise prior, as an inverse Gamma distribution

Details

Currently assumes the covariance function. By default we will use the squared exponential (also called radial basis or Gaussian, though it is not this that gives Gaussian process it's name; any covariance function would do) as the covariance function,

Value

the MCMC output, as constructed by metrop from the mcmc package.

cboettig/nonparametric-bayes documentation built on May 13, 2019, 2:09 p.m.