gp_matern52_cov() [4/6]

template<typename T_x , typename T_s , typename T_l >

Eigen::Matrix< return_type_t< T_x, T_s, T_l >, Eigen::Dynamic, Eigen::Dynamic > stan::math::gp_matern52_cov ( const std::vector< Eigen::Matrix< T_x, Eigen::Dynamic, 1 > > &  x, const T_s &  sigma, const std::vector< T_l > &  length_scale  )

inline

Returns a Matern 5/2 covariance matrix with one input vector with automatic relevance determination (ARD).

\[ k(x, x') = \sigma^2\bigg(1 + \sqrt{5}\sqrt{\sum_{k=1}^{K}\frac{d(x, x')^2}{l_k^2}} + \frac{5}{3} \sqrt{\sum_{k=1}^{K}\frac{d(x, x')^2}{l_k^2}}\bigg) exp\bigg(-\frac{5}{3}\bigg(\sqrt{\sum_{k=1}^K{\frac{d(x, x')^2}{l_k^2}} }\bigg)\bigg) \]

where \( d(x, x') \) is the Euclidean distance.

Template Parameters

T_x	type of elements contained in vector x
T_s	type of element of sigma, the magnitude
T_l	type of elements in vector of length scale

Parameters

x	std::vector of elements that can be used in stan::math::distance
length_scale	length scale
sigma	standard deviation that can be used in stan::math::square

Exceptions

std::domain	error if sigma <= 0, l <= 0, or x is nan or inf
std::invalid_argument	if length scale size != dimension of x

Definition at line 112 of file gp_matern52_cov.hpp.