Compute an oblique target rotation for a loading matrix using a `targetQ`-compatible parameterization and a `k x k` objective.
.oblique_procrustes(
A,
B,
S_r = NULL,
T_init_r = NULL,
eps = 1e-05,
maxit = 1000L,
max_line_search = 10L,
step0 = 1,
normalize = FALSE,
random_starts = 0L,
screen_keep = 2L,
triage_maxit = 25L,
triage_improve_tol = 0
)A named list containing the rotated loadings, transformation matrix, factor correlation matrix, target criterion value, convergence diagnostics, line-search diagnostics, and multi-start summaries.
Numeric matrix. Loading matrix to be rotated.
Numeric matrix. Target loading matrix with the same dimensions as `A`.
Optional numeric `k x k` matrix containing `crossprod(A)`. Supplying this is useful when the same `A` is rotated repeatedly. Ignored when `normalize = TRUE` because the normalized cross-product is different.
Optional numeric `k x k` starting transformation matrix. If `NULL`, the identity matrix is used for the primary start.
Numeric scalar. Convergence tolerance for the projected-gradient norm.
Integer scalar. Maximum number of full projected-gradient updates.
Integer scalar. Maximum number of step-halving attempts after the initial trial step in each line-search phase.
Numeric scalar. Initial step size used in the projected-gradient update.
Logical scalar. If `TRUE`, apply Kaiser normalization before rotation and reverse it after rotation.
Integer scalar. Number of additional random starts.
Integer scalar. Number of screened random starts retained for triage optimization.
Integer scalar. Number of short optimization iterations used in the triage stage.
Numeric scalar. Relative improvement required for a triaged start to be promoted to full optimization.
The rotated loading matrix is defined as `L = A `Phi = t(T) matrix `T` under the oblique normalization constraint `diag(t(T)
The line search is monotone: a candidate is accepted only if it satisfies the sufficient-decrease condition, or, as a numerical fallback, if it at least decreases the objective after all step halvings are exhausted. Non-invertible candidate transformations are rejected rather than evaluated through a pseudo-inverse.
Additional random starts may be requested. To reduce runtime, the solver uses a two-stage strategy for extra starts: cheap objective screening, followed by short triage optimization, followed by full optimization only for starts that improve on the current incumbent by at least `triage_improve_tol`.
The routine is intended for repeated oblique target rotations in workflows such as bootstrap alignment or consensus alignment of exploratory factor solutions across multiply imputed datasets. It follows the same oblique transformation convention as `GPArotation::targetQ()`.
Bernaards, C. A., & Jennrich, R. I. (2005). Gradient projection algorithms and software for arbitrary rotation criteria in factor analysis. *Educational and Psychological Measurement*, 65, 676-696.
Gower, J. C. (1975). Generalized Procrustes analysis. *Psychometrika*, 40, 33-51.