bfa_gs: Bayesian factor analysis used in Gerard and Stephens (2021).

In dcgerard/vicar: Various Ideas for Confounder Adjustment in Regression

Description

Similar to that of Ghosh and Dunson (2009) but with two key differences: (1) the prior is order invariant (though this makes the factors and factor loadings unidentified), and (2) we place hierarchical priors on the uniquenesses (variances).

Usage

bfa_gs(
  Y21,
  Y31,
  Y32,
  k,
  nsamp = 10000,
  burnin = round(nsamp/4),
  thin = 20,
  display_progress = TRUE,
  hetero_factors = TRUE,
  rho_0 = 0.1,
  alpha_0 = 0.1,
  delta_0 = 0.1,
  lambda_0 = 0.1,
  nu_0 = 1,
  beta_0 = 1,
  eta_0 = 1,
  tau_0 = 1
)

Arguments

Y21	Top left of matrix.
Y31	Bottom left of matrix.
Y32	Top right of matrix.
k	The rank of the mean matrix.
nsamp	A positive integer. The number of samples to draw.
burnin	A positive integer. The number of early samples to discard.
thin	A positive integer. We will same the updates of Y22 every thin iteration of the Gibbs sampler.
display_progress	A logical. Should we print a text progress bar to keep track of the Gibbs sampler (TRUE) or not (FALSE)?
hetero_factors	A logical. Should we assign colum-specific variances for the factors (TRUE) or not (FALSE)?
rho_0	The prior sample size for column-specific the precisions.
alpha_0	The prior sample size for the mean of the column-specific precisions.
delta_0	The prior sample size of the column-specific precisions of the factors.
lambda_0	The prior sample size of the mean of the column-specific precisions of the factors.
nu_0	The prior mean of the mean of the column-specific precisions of the factors.
beta_0	The prior mean of the mean of the column-specific precisions.
eta_0	The prior sample size of the expanded parameters.
tau_0	The prior mean of of the expanded parameters.

Author(s)

David Gerard

References

• Gerard, David, and Matthew Stephens. 2021. "Unifying and Generalizing Methods for Removing Unwanted Variation Based on Negative Controls." Statistica Sinica, 31(3), 1145-1166. doi: 10.5705/ss.202018.0345

• Ghosh, J. and Dunson, D.B., 2009. "Default prior distributions and efficient posterior computation in Bayesian factor analysis." Journal of Computational and Graphical Statistics, 18(2), pp.306-320. doi: 10.1198/jcgs.2009.07145

See Also

gdfa for the slower R implementation.

dcgerard/vicar documentation built on July 7, 2021, 1:08 p.m.