solver

int stan::math::laplace_options_base::solver {1}

Which linear solver to use inside the Newton step.

This selects how the system matrix B is formed and factorized. For details, see equation 1 in: https://arxiv.org/pdf/2306.14976

1. Hessian-root Cholesky: form B = I + W_r * Sigma * W_r where W_r = sqrt(W) and W is the negative Hessian of the log-likelihood w.r.t. theta. This uses either a diagonal (hessian_block_size == 1) or block-diagonal (hessian_block_size > 1) approximation of W.
2. Covariance-root Cholesky: precompute K_root such that Sigma = K_root * K_root^T and form B = I + K_root^T * W * K_root.
3. General LU: form B = I + Sigma * W and factorize with LU.

Definition at line 49 of file laplace_marginal_density_estimator.hpp.