dot-compute_dhat_diag: Compute the Derivative of the Tuning Hat Matrix Diagonal

In lgspline: Lagrangian Multiplier Smoothing Splines for Smooth Function Estimation

Compute the Derivative of the Tuning Hat Matrix Diagonal

Description

Computes d h_{ii} / d\lambda_1 partition-wise for the exact leave-one-out tuning criterion.

Usage

.compute_dhat_diag(X, G, dG_dlambda, A, AGAInv, K, p_expansions)

Arguments

X	List of partition-specific design matrices.
G	List of partition-specific \mathbf{G}_k matrices.
dG_dlambda	List of derivatives d\mathbf{G}_k / d\lambda_1.
A	Constraint matrix.
AGAInv	Inverse of \mathbf{A}^{\top}\mathbf{G}\mathbf{A}.
K	Number of interior knots.
p_expansions	Number of columns per partition.

Value

List with dh_diag and dC_list, both partition-wise.

lgspline documentation built on May 8, 2026, 5:07 p.m.