ebpm_pois_sgp: Poisson Sparse Gaussian Process

In DongyueXie/smashrgen: Empirical Bayes smoothers

Poisson Sparse Gaussian Process

Description

Poisson Sparse Gaussian Process

Usage

ebpm_pois_sgp(
  x,
  s = 1,
  g_init = NULL,
  fix_g = FALSE,
  q_init = NULL,
  control = list()
)

Arguments

x	A vector of Poisson observations.
s	A vector of scaling factors for Poisson observations
g_init	a list of prior parameters: X_ind, kernel_param, mu, kernel_func
fix_g	a boolean or a vector of boolean of length 3. If TRUE, set the corresponding prior at g_init.
q_init	a list of init values of posterior: mean_log_ind, v_log_ind. These are the posterior mean and var of the f at the inducing points.
control	a list of other parameters passed to the 'pois_sgp' function.

Details

The model is

x_{i}\sim \text{Poisson}(s_i\exp(\mu+f_{i})),

f\sim N(0,\Sigma(\theta)).

Value

a list of posterior, log_likelihood, fitted_g

DongyueXie/smashrgen documentation built on Jan. 14, 2024, 5:30 a.m.