1Blurb:: 2Determine how the final covariance matrix is computed 3Description:: 4\c covariance 5(NL2SOL's \c covreq) specifies whether and how NL2SOL computes 6a final covariance matrix. 7 8The desired covariance approximation: 9\li 0 = default = none 10\li 1 or -1 ==> \f$\sigma^2 H^{-1} J^T J H^{-1}\f$ 11\li 2 or -2 ==> \f$\sigma^2 H^{-1}\f$ 12\li 3 or -3 ==> \f$\sigma^2 (J^T J)^{-1}\f$ 13\li Negative values ==> estimate the final Hessian H by finite 14differences of function values only (using \c fd_hessian_step_size) 15\li Positive values ==> differences of gradients (using 16\c fd_hessian_step_size) 17 18 19Topics:: 20Examples:: 21Theory:: 22Faq:: 23See_Also:: 24