logprior_W: log-prior of W

In awstringer1/casecrossover: Approximate Bayesian Inference for Case Crossover Models

Description

These functions calculate priors and posteriors of the latent Gaussian variables.

Usage

logprior_W(W, model_data, theta = NULL, Q = NULL)

log_posterior_W(W, theta, model_data, Q = NULL)

grad_log_posterior_W(W, theta, model_data, Q = NULL)

hessian_log_posterior_W(
  W,
  theta = NULL,
  Q = NULL,
  model_data,
  structure = NULL
)

Arguments

W	Value of W = (delta,gamma,beta) to calculate the prior/posterior at. The order is as it appears in the appendix of the paper- (delta_1_1,...,delta_n_Jn,gamma_1,...,gamma_M,beta_1,...,beta_p). The order of gamma and beta is the same as they are listed in model_data$model_elements.
model_data	A ccmodeldata object returned by model_setup().
theta	Value of the hyperparameter vector theta. The Q matrix depends on this. Can leave as NULL if you're passing in the Q matrix
Q	The Q-matrix as returned by Q_matrix(model). If not provided, will be calculated.
structure	Optional. Pass in the sparse structure of the hessian of the log-likelihood to save computing time. See hessian_log_likelihood_structure().

awstringer1/casecrossover documentation built on March 11, 2021, 4:41 a.m.