中文摘要：

大量经验研究发现宏观基本面显著影响收益率曲线及其期限结构特征。本文提出一种包含低频宏观因子的混频Nelson—Siegel模型并基于中国国债收益率及宏观经济变量展开讨论。研究结果表明：混频模型可以改进同频模型拟合效果并较好刻画出具有明确经济含义的期限结构水平、斜率和曲度因子，说明混频模型在拟合中国国债收益率曲线方面的适应性。此外，我们发现水平因子对通货膨胀有明显作用，尤为重要的是，曲度因子受到GDP潜在因子显著的正向影响；最后，基于方差分解可以发现通胀因子主要作用于水平因子及收益率曲线长端，而GDP潜在因子则对曲度因子及中期收益率的贡献较大。

英文摘要：

Many empirical studies find that macroeconomic fundamentals have also significant influence in the yield curve. This paper proposes and estimates a mixed - frequency Nelson - Siegel term structure model with the Chinese Treasure yields and c data. The results show that the mixed -frequency model can improve the in - sample fit of the yield curve compared with the basic Nelson - Siegel model. The term structural （ the level, slope and curvature） factor implied the clear economic implications can be extracted by the model. In addition, the model can identify that the levels factor significantly affect the inflation. More importantly, the curvature factor is affected by and GDP latent factor. Finally, according to the variance decomposition, we find that the inflation plays important rule in level factor and long term yield curve. The latent factor of GDP mainly contribution to curvature factor and short or middle term yield curve.