序贯观察与选择问题截止阀策略的一个改进

An extension on cut-off rules of sequential observation and selection problem

截止阀策略是解决序贯观察与选择问题的有效方法.在多次应用的问题情境下,通过减少截止阀值来降低失败的概率,进而实现平均收益的最大化,代价是减少了观察选项的数量,增加了决策的不确定性.本研究围绕降低失败的概率进行,从降低标竿而非减少截止阀值入手,采用均值标杆改进了最大值标竿的截止阀策略.基于仿真实验的结果显示,改进后的策略具有更好的平均收益和更低的失败概率,同时,还证明了20%就是均值标竿的近似最优解.

Cut-off rules specialized in sequential observation and selection problem. The established cutoff rules decreased the cutoff value for low risk in repetitive applications context. Furthermore, the decreasing of cutoff value can maximize the average profit which cost is reducing the number of observation and enhancing the uncertainty of decision. In this paper, reducing the benchmark is more efficient than decreasing cutoff value in risk aversion. The established cut-off rule had been revised and the results, based on the simulations, indicate that the revised cut-off rule is risk aversion and has better average profit. The prove process that the 20 percent is the approximate optimal proportion bring this paper to a close.