estimate_noise_nn {GPCERF}R Documentation

Estimate the standard deviation (noise) of the nugget term in nnGP

Description

Estimates the standard deviations of the nugget term (noise) in nnGP by calculating the standard deviations of the residuals.

Usage

estimate_noise_nn(
  hyperparam,
  w_obs,
  GPS_obs,
  y_obs,
  n_neighbor,
  nthread,
  kernel_fn = function(x) exp(-x^2)
)

Arguments

hyperparam

A vector of hyper-parameter values.

w_obs

A vector of observed exposure levels.

GPS_obs

A vector of estimated GPS evaluated at the observed exposure levels.

y_obs

A vector of observed outcomes.

n_neighbor

A number of nearest neighbors on one side.

nthread

A number of cores used in the estimation.

kernel_fn

The covariance function of the GP.

Value

A scalar of estimated standard deviation of the nugget term in nnGP.


[Package GPCERF version 0.2.4 Index]