作者: Douglas M. Bates , Mary J Lindstrom , Grace Wahba , Brian S Yandell
DOI: 10.1080/03610918708812590
关键词:
摘要: These Fortran-77 subroutines provide building blocks for Generalized Cross-Validation (GCV) (Craven and Wahba, 1979) calculations in data analysis smoothing including ridge regression (Golub, Heath, 1979), thin plate splines (Wahba Wendelberger, 1980), deconvolution (Wahba, 1982d), of generalized linear models (O'sullivan, Yandell Raynor 1986, Green 1984 1985), ill-posed problems (Nychka et al., 1984, O'sullivan 1985). We present some the types which GCV is a useful method choosing or regularization parameter we describe structure subroutines.Ridge Regression: A familiar example X problem write.