Modelling the Term Structure of Interest Rates and Volatility in China

  • Songzhuo Li

Student thesis: Doctoral ThesisPhD

Abstract

The purpose of this study is to examine the dynamic behaviour of the Chinese yield curve and the short rate volatility. It consists of an element of literature reviews on the term structure of interest rates and three empirical chapters. The empirical studies discuss the dynamics of Chinese yield curve, the interactions between Chinese yield curve and economy and the volatility in the Chinese short-term interest rate. First, we employ the Fourier model to estimate the term structure of Chinese interest rates, following Moreno, Novales and Platania (2013). The Fourier model is an extension of Vasicek model by imposing a Fourier series to describe the long-run mean. The Fourier model provides better approximation and prediction of the dynamics of Chinese yield curve than the Vasicek model, especially on the short end. We conclude that the Fourier assumption does help to capture the volatility of Chinese yield curves, as the Chinese yield curve is found to behave cyclically. Second, we construct and estimate the Nelson-Siegel form macro-finance model based on Chinese market, following Diebold, Rudebusch and Aruoba (2006). Bidirectional causality is found, however the yield curve effect on the macroeconomy is relatively weak compared to the reverse influence. In the long-term horizon, both the inflation rate and real activity as approximate by industrial production, can explain more than 30 percent of the variation of yield curve. Finally, we examine the dynamic behaviour of Chinese short rate in frame of Conley, Hansen, Luttmer and Scheinkman (1997). Four one-factor diffusion models and four Markov regime-switching extensions are compared. We find that incorporating regime-switching can largely improve in-sample fitting to data and also help to capture the volatility clustering and fatter tail. The nonlinearity of drift term seems of importance on capturing the movement of Chinese short rate.
Date of Award22 Dec 2016
Original languageEnglish
Awarding Institution
  • University of Bath
SupervisorVito Polito (Supervisor) & Christos Ioannidis (Supervisor)

Cite this

'