预调试
CVAR公司
杠杆(统计)
经济
数学优化
预期短缺
资产配置
资产(计算机安全)
计量经济学
风险度量
时间一致性
计算
计算机科学
数理经济学
微观经济学
数学
精算学
风险管理
财务
文件夹
机器学习
计算机安全
算法
摘要
We formulate the multiperiod, time consistent mean-CVAR (conditional value at risk) asset allocation problem in a form amenable to numerical computation. Our numerical algorithm can impose realistic constraints such as no shorting, no leverage, and discrete rebalancing. We focus on long term (i.e., 30 year) strategies, which would be typical of an investor in a defined contribution pension plan. A comparison with precommitment mean-CVAR strategies shows that adding the time consistent constraint compares unfavorably with the pure precommitment strategy. Since the precommitment strategy computed at time zero is identical to a time consistent strategy based on an alternative objective function, the time zero precommitment mean-CVAR strategy is implementable in this case. Hence it would seem that there is little to be gained from enforcing time consistency.
科研通智能强力驱动
Strongly Powered by AbleSci AI