parDA_theta_MH: Update parameters from the model in the MCMC algorithm

In mapn/DAbayes: Bayesian Statistical Model for Climate-Change Detection and Attribution

Description

This function updates parameters for the Bayesian statistical model. gamma is updated with a discrete uniform prior; beta is updated with noninformative prior 1 and random-walk proposal; log of sigma (variance in W) is updated with normal prior and random-walk proposal; lambda is updated with normal prior and random-walk proposal. This function is used in parallel computing of the MCMC algorithm.

Usage

parDA_theta_MH(outW, theta, prior_theta, ADtheta, loglik_old, chisq_old, N, Lj,
  tol = 1e-10)

Arguments

outW	a 7 by 1 list containing precomputed quantities associated with W
theta	a list of 3 elements contained in parameters in MCMC algorithm
prior_theta	quantities (parameters) in prior dist.
ADtheta	a 6 by 1 list containing quantities to adjust mean and covariance in proposal dist
loglik_old	a real number representing likelihood for given parameters
chisq_old	a real number representing the quadratic form
N	number of temperature ensemble members
Lj	an m by 1 vector containing the number of runs for each forcing scenario
tol	a very small value to avoid numerical instability

Value

a list of updated parameters, loglikelihood, chisq statistics (quadratic form), and prior

Author(s)

Pulong Ma <[e-mail]>

See Also

DA_theta_MH

mapn/DAbayes documentation built on May 21, 2019, 11:26 a.m.