| 英文摘要 |
This paper investigates the relationship between interest rates and the price level using the Principal Components method, Nelson-Siegel model, and Diebold-Li model to extract implicit information from the term structure of interest rates. Specifically, the interest rates of the highest credit-rating corporate debt securities from 2005 to 2020 are used to investigate the sectoral price indices from tradable and non-tradable sectors to prevent potential aggregation bias. All three approaches help explain the sectoral price movements, particularly the Diebold-Li method. Estimation results for tradable goods, non-tradable goods, and the aggregate price index vary significantly. In other words, the findings imply that it is necessary to apply sectoral price indices instead of an aggregate price index. Notably, model uncertainty issues with the price index prediction arise when we path of the uncertainty index. We, therefore, utilize a moving-window regression to illustrate the problems of uncertainty and find that the prediction results change because of the uncertainty arising in 2007, 2010, 2015, and 2020. Finally, the term structure of interest rates demonstrates a non-trivial ability of reflect sectoral price changes, despite uncertainties related to the price index. |