When Restrictive Constraints Are Nonbinding: Illustrations and Implications
通过数值例子说明，在凸可行集问题中，限制性约束不一定在最优解处起作用，并讨论了这一发现对公共政策分析、计量估计和求解算法的启示。
For convex and concave mathematical programs restrictive constraints (i.e., their deletion would change the optimum) will always be binding at the optimum, and vice versa. Less well‐known is the fact that this property does not hold more generally, even for problems with convex feasible sets. This paper demonstrates the latter fact using numerical illustrations of common classes of problems. It then discusses the implications for public policy analysis, econometric estimation, and solution algorithms.