ss_ranef_mv: ss_ranef_mv

In josue-rodriguez/SSranef: Mixed-effects Models with Spike-and-Slab Priors

ss_ranef_mv

Description

ss_ranef_mv() fits a multivariate mixed-effects model with random intercepts and random slopes for two outcomes. A spike-and-slab prior is placed on the random slopes of both outcomes.

Usage

ss_ranef_mv(
  Y,
  X,
  unit,
  burnin = 1000,
  iter = 1000,
  chains = 4,
  priors = NULL,
  vars2monitor = c("B", "theta", "gamma1", "gamma2", "sigma", "Tau", "rb", "rw")
)

Arguments

Y	A two column matrix containing the outcomes of interest.
X	A vector containing the predictor variable.
unit	A vector of containing a unique identifier for each row in Y.
burnin	The number of iterations to use as burnin. Defaults to 1000.
iter	The number of iterations to use for estimating the parameters. Defaults to 1000.
chains	The number of MCMC chains to use. Defaults to 4.
priors	A named list to specify priors. Defaults to NULL. See README for details.
vars2monitor	A vector containing the names of which parameters to monitor. See details below.

Details

The parameters that can be tracked are:

• B: The fixed effects.

  • B_1_1 and B_1_2 corresponds to the fixed effect intercept and slope for the first outcome, respectively.

  • B_2_1 and B_2_2 corresponds to the fixed effect intercept and slope for the second outcome, respectively.

• gamma1: The inclusion indicators for the slope random effects of the first outcome.

• gamma2: The inclusion indicators for the slope random effects of the second outcome.

• rb: The between-unit correlations, or the correlations between the random effects.

  • rb_1_2: Correlation between random intercept and random slope for the first outcome

  • rb_1_3: Correlation between random intercept for the first outcome and random intercept for the second outcome

  • rb_1_4: Correlation between random intercept for the first outcome and random slope for the second outcome

  • rb_2_3: Correlation between random slope for the first outcome and random intercept for the second outcome

  • rb_2_4: Correlation between random slope for the first outcome and random slope for the second outcome

  • rb_3_4: Correlation between random intercept and random slope for the second outcome

• rw: The correlation between the residual standard deviations, sigma_1 and sigma_2

• sigma: The residual standard deviations sigma_1 (first outcome) and sigma_2 (second outcome)

• Tau: The (co-)variances for the random effects

  • Tau_1_1: Variance for the random intercept of the first outcome

  • Tau_2_2: Variance for the random slope of the first outcome

  • Tau_3_3: Variance for the random intercept of the second outcome

  • Tau_4_4: Variance for the random slope of the second outcome

• theta: The random effects each outcome. The names are formatted as theta_unitID_randomEffect. E.g., theta_3_2 would correspond to the random intercept of the 2nd outcome for the 3rd unit.

josue-rodriguez/SSranef documentation built on March 23, 2022, 3:27 p.m.