Sample AR(1) parameter for dynamic additive effects
Usage
sample_rho_ab_cpp(
a_mat,
b_mat,
sigma_ab,
rho_current,
symmetric,
prior_mean = 0,
prior_sd = -1
)Arguments
- a_mat
Matrix of row effects (n x T)
- b_mat
Matrix of column effects (n x T)
- sigma_ab
Innovation standard deviation
- rho_current
Current value of rho
- symmetric
Whether the network is symmetric
- prior_mean
Prior mean for rho. Used only when
prior_sd > 0.- prior_sd
Prior SD for rho.
prior_sd < 0(the default) selects a Jeffreys-like flat prior; a positive value switches to a truncated Normal(prior_mean, prior_sd^2) prior.