基于LAD-Lasso方法的利率期限结构拟合中的节点选择

Knot selection of estimating the term structure with cubit spline function based on LAD-Lasso criterion

将LAD-Lasso方法引入到三次样条函数中,通过LAD-Lasso来进行变量选择,确定样条函数的节点数量和位置,同时估计参数,构建模型来拟合上海证券交易所交易的国债利率期限结构.样本外预测结果显示,与传统的方法相比,新方法可以有效地选择合适的模型,增加参数估计的稳健性,简化计算步骤,提高预测的精度,增强期限结构定价的准确度.

LAD-Lasso method was employed to estimate the Shanghai stock exchange term structure of interest rates with cubic spline function based on least absolute deviations （LAD） criterion. The forecasting power of the LAD-Lasso method based on LAD criterion was compared with the traditional knot settings. The out-of-sample forecasting results show that the model based on the new method improves the pricing precision of the bonds.