Full pairwise iteration — cpp_multiThread_completePairwise • lavaan.pl

Evaluate negative log pairwise likelihood or gradient of the complete pool of pairs. Used by external optimisers. Multithreading options via RcppParallel.

Usage

cpp_multiThread_completePairwise(
  N,
  C_VEC,
  CONSTRMAT,
  CONSTRLOGSD,
  LLC,
  THETA,
  FREQ,
  CORRFLAG,
  NTHR,
  NLOAD,
  NCORR,
  NVAR,
  GRFLAG = 0L,
  SILENTFLAG = 1L
)

Arguments

N: Number of observations
C_VEC: Vector containing the number of categories for each item
CONSTRMAT: \(p*q\)-dimensional matrix. Elements set to NA refers to free loading parameters. Elements set to numerical values denote fixed values constraints.
CONSTRLOGSD: \(q\)-dimensional vector. Elements set to NA refers to free latent log standard deviations parameters. Elements set to numerical values denote fixed values constraints.
LLC: Linear loadings constraints. Expects a list of constraints. See fit_plFA documentation.
THETA: Parameter vector
FREQ: Frequency table
CORRFLAG: TRUE to estimate latent correlations. 0 for orthogonal latent factors.
NTHR: Number of thresholds parameters.
NLOAD: Number of free loadings parameters
NCORR: Number of free latent correlations parameters.
NVAR: Number of free latent variance parameters.
GRFLAG: 0 to only compute the likelihood. 1 to also compute the gradient.
SILENTFLAG: optional for verbose output