Post-process Stack FA output (nutrition tutorial)

Implements the post-processing in the nutrition case study: apply orthogonal Procrustes (OP) to posterior loadings, form SigmaPhi = Phi Phi^T, estimate marginal covariance by averaging Sigma draws, and estimate Psi by averaging Sigma - Lambda Lambda^T over MCMC iterations.

Usage

postprocess_stack_fa(fit, S)

Arguments

fit

A bifa_fit object from fit_stack_fa() or a raw MSFA::sp_fa() fit.

S

Number of studies (used only to replicate marginal covariance into a list).

Value

A list with Phi, SigmaPhi, Psi, SigmaMarginal.