run_sampler_binder: Run BINDER model using prepared data.

In ptrcksn/BINDER: BayesIan gene regulatory Networks inferreD via gene coExpression and compaRative genomics

Description

Run BINDER model using prepared data.

Usage

run_sampler_binder(prepared_data, hyperparams = NULL, inits = NULL,
  n_chains = 1, n_draws = 1000, burn_in = 0, thin = 0, seed = 1)

Arguments

prepared_data	A named list generated by prepare_data function comprising model input data.
hyperparams	A named list of user-defined hyperparameters; this list should contain some or all of the following named values: mu_zeta (prior mean for auxiliary stratum intercept (normal distribution)), sigma_zeta (prior variance for auxiliary stratum intercept (normal distribution)), mu_tau_ME (prior mean for auxiliary stratum ME coefficient (normal distribution)), sigma_tau_ME (prior variance for auxiliary stratum ME coefficient (normal distribution)), mu_tau_PE (prior mean for auxiliary stratum PE coefficient (normal distribution)), sigma_tau_PE (prior variance for auxiliary stratum PE coefficient (normal distribution)), alpha_phi (prior shape parameter for phi (inverse gamma distribution)), beta_phi (prior scale parameter for phi (inverse gamma distribution)), alpha_psi_CM (prior shape parameter for psi_CM (inverse gamma distribution)), beta_psi_CM (prior scale parameter for psi_CM (inverse gamma distribution)), alpha_psi_CP (prior shape parameter for psi_CP (inverse gamma distribution)), beta_psi_CP (prior scale parameter for psi_CP (inverse gamma distribution))); for a given hyperparameter, default values are used if no user-defined value is provided to the function call (mu_zeta := 0, sigma_zeta := 1, mu_tau_ME := 0, sigma_tau_ME := 1, mu_tau_PE := 0, sigma_tau_PE := 1, alpha_phi := 1.5, beta_phi := 1.5, alpha_psi_CM := 1.5, beta_psi_CM := 1.5, alpha_psi_CP := 1.5, beta_psi_CP := 1.5).
inits	A named list of user-defined initial values to begin each MCMC chain (each element of this list should comprise a vector of length equal to the number of chains and each element of the vector should correspond to the initial value for that parameter for each chain); this list should contain some or all of the following named values: zeta_init (defaults to randomly generated value from prior normal distribution for zeta), tau_ME_init (defaults to randomly generated value from prior normal distribution for tau_ME), tau_PE_init (defaults to randomly generated value from prior normal distribution for tau_PE), phi_init (defaults to randomly generated value from prior inverse-gamma distribution for phi), psi_CM (defaults to randomly generated value from prior inverse-gamma distribution for psi_CM), psi_CP (defaults to randomly generated value from prior inverse-gamma distribution for psi_CP).
n_chains	Number of MCMC chains to run; defaults to 1.
n_draws	Number of MCMC draws to run; defaults to 1000.
burn_in	Number of MCMC draws to discard from the beginning of each chain; defaults to 0.
thin	Number of MCMC draws to discard between each accepted MCMC draw for each chain; defaults to 0.
seed	Seed for random number generation; defaults to 1.

Value

A list containing the MCMC draws for each model parameter for each chain.

ptrcksn/BINDER documentation built on Nov. 5, 2019, 1:56 a.m.