corr_matrix_free() [1/2]

template<typename T , require_eigen_t< T > * = nullptr>

Eigen::Matrix< value_type_t< T >, Eigen::Dynamic, 1 > stan::math::corr_matrix_free

(

const T &

y

)

Return the vector of unconstrained partial correlations that define the specified correlation matrix when transformed.

The constraining transform is defined as for corr_matrix_constrain(Matrix, size_t). The inverse transform in this function is simpler in that it only needs to compute the $k \choose 2$ partial correlations and then free those.

Template Parameters

T	type of the matrix (must be derived from Eigen::MatrixBase)

Parameters

y	The correlation matrix to free.

Returns

Vector of unconstrained values that produce the specified correlation matrix when transformed.

Exceptions

std::domain_error	if the correlation matrix has no elements or is not a square matrix.
std::runtime_error	if the correlation matrix cannot be factorized by factor_cov_matrix() or if the sds returned by factor_cov_matrix() on log scale are unconstrained.

Definition at line 34 of file corr_matrix_free.hpp.