Hyperparameter and MCMC control settings for bcaia

Collects the fixed hyperparameters of the model together with the Metropolis-Hastings tuning constants into a single list. All values have defaults matching those used in the simulation studies of Zhang, Patnode and Lee; override individual entries by passing them as named arguments.

Usage

bcaia_control(
  a_phi = 1/20,
  a_tau = 0.1,
  b_tau = NULL,
  a_sig = 3,
  b_sig = 3,
  u2_beta = 1,
  Lr = 30,
  L_alpha = 35,
  a_psi_r = 1,
  a_psi_alpha = 1,
  a_w = 5,
  b_w = 5,
  a_w_alpha = 5,
  b_w_alpha = 5,
  ur2 = 1,
  acc_tar = 0.234
)

Arguments

a_phi, a_tau, b_tau

Dirichlet-Horseshoe hyperparameters for the factor loadings. Defaults a_tau = 0.1, b_tau = 1/J (set internally from the data when NULL) and a_phi = 1/20.

a_sig, b_sig

Inverse-gamma prior parameters for the idiosyncratic variance \(\sigma^2\). Default 3 and 3.

u2_beta

Prior variance of the mean-regression coefficients \(\beta_{jp}\). Default 1.

Lr, L_alpha

Stick-breaking truncation levels for the size factor \(r\) and baseline abundance \(\alpha\). Defaults 30 and 35.

a_psi_r, a_psi_alpha

Dirichlet-process concentration parameters for \(r\) and \(\alpha\). Default 1 (the subject-indexed models in the paper use 3).

a_w, b_w, a_w_alpha, b_w_alpha

Beta prior parameters for the inner mixture weights. Default 5 each.

ur2

Prior variance of the size factor \(r\). Default 1.

acc_tar

Target acceptance rate for the adaptive Metropolis-Hastings updates. Default 0.234.

Value

A named list of class "bcaia_control".

Examples

ctrl <- bcaia_control(a_psi_r = 3, a_psi_alpha = 3)