论文部分内容阅读
通过对比国内外利率期限结构静态估计模型的优劣,分析节点数目变化和定位改进B样条函数对利率期限结构静态估计的误差,构建最小化定价误差的节点组合布局搜索程序,并引入负指数平滑立方L1样条优化模型,将误差函数最小化结构从平方和最小化转化为误差距离最小化,权衡拟合误差绝对距离最小化与贴现函数波动性约束,克服B样条函数对节点数目与定位的人工干预和放宽对贴现函数的二阶平滑要求,保留B样条函数刻画中长期利率波动趋势的优势,增强对短期利率波动结构突变的估计和预测能力,提高定价精确度和缓解利率期限结构曲线的过度波动问题.
By comparing the advantages and disadvantages of the static estimation model of the term structure of interest rate at home and abroad, the error of the static estimation of the term structure of the interest rate is analyzed by the change of the number of nodes and the location improved B-spline function, and the node combination layout search program of minimizing the pricing error is constructed. Smooth Cubic L1 spline optimization model, the minimization of error function from the square summation minimization error distance to minimize the absolute error of the fitting error minimization and discount function volatility constraints, to overcome the B-spline function of the number of nodes and Positioning artificial intervention and relaxation of second-order smoothing requirements for discount functions, retaining the advantages of B-spline function to depict the trend of long-term interest rate volatility, enhancing the ability to estimate and predict sudden changes in short-term interest rate structure, improving pricing accuracy and easing interest rates Excessive fluctuation of structural curve.