![]() Statisticians often use least-squares minimization in the context of linear or nonlinear least-squares (LS) regression. In SAS Viya, you can use the FDDSOLVE subroutine, which is a newer finite-difference function that has a syntax similar to the NLPFDD subroutine. The NLPFDD subroutine has been in SAS/IML software since SAS version 6. ![]() This article expands on that statement and shows that you can use the SAS/IML matrix language to implement the Gauss-Newton method by writing only a few lines of code. This product is used in some numerical methods, such as the Gauss-Newton algorithm, to minimize the value of a vector-valued function." States that the third output argument of the NLPFDD subroutine "contains the matrix product J`*J, where J is the Jacobian matrix. ![]() The article uses the NLPFDD subroutine in SAS/IML to compute the finite-difference derivatives. A previous article showed how to use SAS to compute finite-difference derivatives of smooth vector-valued multivariate functions.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |