Stepwise forward and backward selection of variables using linear models

Abstract There a number of interesting variable selection methods available beside the regular forward selection and stepwise selection methods. Such approaches include LASSO (Least Absolute Shrinkage and Selection Operator), least angle regression (

cholesky分解是用于将一个对称正定的矩阵分解为一个上三角与一个下三角矩阵乘积的形式，可运用于回归系数求解，线性方程组求解中 逐步回归是在回归分析的建模过程中用于变量筛选的有效方法，这里同时进行了向前和向后回归，效果更加显著(Cholesky decomposition is used to decompose a symmetric positive definite matrix into an upper triangle and a lower triangular matrix

包括LARS的经典文章和实现代码（MATLAB）(Abstract There are a number of interesting variable selection methods available beside the regular forward selection and stepwise selection methods. Such approaches include lasso (Least Absolute Shrinkage and Selection Operat