log_posterior_theta: Log-posterior approximation for theta and sigma

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

Description

Compute the log-posterior for theta, the log-precision, and sigma, the standard deviation, of the RW2 model components.

Usage

log_posterior_theta(theta, W, model_data, hessian_structure = NULL, Q = NULL)

log_posterior_sigma(sigma, W, model_data, hessian_structure = NULL, Q = NULL)

Arguments

theta	Vector of log-precisions at which to evaluate the log-posterior. theta = -2*log(sigma).
W	The mode of the log-posterior of W|theta,y for this theta. Computed outside and passed in.
model_data	a ccmodeldata object created using model_setup()
hessian_structure	Optional, sparse structure of the Hessian created using hessian_log_likelihood_structure()
Q	Optional, Q matrix evaluated at this theta. Will be created if not supplied.
sigma	Vector of standard deviations at which to evaluate the log-posterior. sigma = exp(-.5 * theta)

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