主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2012, Vol. 20 ›› Issue (6): 28-34.

• 论文 • 上一篇    下一篇

国债市场利率期限结构波动的对偶变换建模

吴泽福   

  1. 华侨大学工商管理学院,福建 泉州 362021
  • 收稿日期:2010-11-08 修回日期:2012-09-17 出版日期:2012-12-29 发布日期:2012-12-28
  • 基金资助:
    国家自然科学基金资助项目(70573033);教育部规划基金(12JA790147);泉州市哲社规划项目(2012Y04)

Dual Tranformation Modelling on Term Structure of Interest Rate in T-bill Maket

WU Ze-fu   

  1. College of Business Administration,Huaqiao University,Quanzhou 362021,China
  • Received:2010-11-08 Revised:2012-09-17 Online:2012-12-29 Published:2012-12-28

摘要: 本文针对传统利率期限结构拟合曲线存在过度波动问题,构建定价误差绝对距离和波动曲率双重最优化模型,借助对偶几何程序转换为在线性约束区域内的绝对距离最小化问题,并运用负指数平滑立方L1样条和计算几何逼近算法求解模型参数,通过负指数立方L1样条、NSS模型和B样条进行样本内拟合与样本外预测能力的比较,证实负指数立方L1平滑样条对利率期限结构波动的定价精确度、结构性拟合和样本外预测能力均有明显的优势,丰富了国债市场利率期限结构波动与定价的理论基础和研究方法。

关键词: 国债市场, 利率期限结构, 波动建模, 对偶变换

Abstract: In order to be against excessive volatility of traditional fitting curve for term structure of interest rate(TSIR), optimization model on minimizing function value with absolute price error and volatility rate is crcated,which is transformed into minimization programming of absolute distance within linearity constraint,and the model’s parameters are figured through negative exponential smoothness cubic L1-spline and numerical geometric’s approximative arithmetic.Negative exponential smoothness cubic L1-spline displays strong capability in pricing accuracy,volatility structure fitting and multi-step forecast,with comparison to NSS model and B-spline on fitting and forecast capability.The result enriches theorical basis and research method on TSIR’s volatility and valuation in T-bill market.

Key words: T-bill market, term structure of interest rate, volatility modelling, dual transformation

中图分类号: