Variable Selection Via Subtle Uprooting

摘要: This article proposes a variable selection method termed “subtle uprooting” for linear regression. In this proposal, is formulated into single optimization problem by approximating cardinality involved in the information criterion with smooth function. A technical maneuver then employed to enforce sparsity of parameter estimates while maintaining smoothness objective To solve resulting nonconvex problem, modified Broyden–Fletcher–Goldfarb–Shanno (BFGS) algorithm established global and super-linear convergence adopted. Both simulated experiments an empirical example are provided assessment illustration. Supplementary materials available online.