dot-solver_recover_active_multipliers: Recover Active-Constraint Multipliers from Transformed OLS

In lgspline: Lagrangian Multiplier Smoothing Splines for Smooth Function Estimation

Recover Active-Constraint Multipliers from Transformed OLS

Description

Given the transformed regression pair (\mathbf{y}^{*}, \mathbf{X}^{*}), extracts the implied KKT multipliers for the appended active inequalities.

Usage

.solver_recover_active_multipliers(
  X_star,
  y_star,
  n_eq_orig,
  K,
  tol = sqrt(.Machine$double.eps),
  ineq_sign = -1,
  parallel_qr = FALSE,
  cl = NULL
)

Arguments

X_star	Transformed constraint matrix.
y_star	Transformed right-hand side.
n_eq_orig	Number of leading equality constraints in X_star.
K	Integer; number of interior knots.
tol	Numeric tolerance.
ineq_sign	Numeric sign mapping from transformed-OLS coefficients to KKT multipliers. For \mathbf{C}^{\top}\beta \ge c, use -1.

Value

List with multipliers, drop, and the full transformed-OLS coefficients.

lgspline documentation built on Aug. 5, 2026, 1:10 a.m.