The Empirical Tests Of The Expectations Of Term Structure Of Interest Rate Based On SHIBOR

Interest rate has many important functions in the economy, it has been seen a sensitive economic levers in the economic operation. In macroeconomics, monetary policy authorities usually saw interest rate as a flexible and effective monetary policy tools, central banks often adjust the allocation of resources, money supply and demand, and control inflation by adjusting the size of the interest rate. But using the interest rate to meet our expectations effective is failed, so why not use interest rate in our ideal? The function of interest rate is not work in our country, or we do not have the economic conditions of adjusting the interest rates, or the institutional management of the economy is not sufficiently standardized. Therefore, in reality, the monetary authorities and a considerable portion of our experts conducted in-depth research in both of theory and application, through with hard explorations but we still do not achieve the desired results.The level of real interest rates determine the economic agent’s consumption and investment decisions, the Fisher effect shows that the nominal interest rate compose the real interest rate and the information of expected inflation, it is already included compensation for the risk of inflation rates, thus bond yields depicted consist of different kinds of deadline risk-free also called yield curve the term structure of macro-theoretical analysis and decision-making has a very important role. A lot of economic data suggest that the term structure of interest rates can fully reflect the true information economy. Through the relationships of current long-term interest rates, and market participants can determine the future trend of interest rates, and determine the future trend of economic growth and inflation based on the slope of the yield curve (short and long term interest rate).Many foreign experts and scholars founded the "anticipate mystery" in the term structure of interest rates when they are using the data of United States for research."Anticipate mystery" is expected to be a mystery paradox refers to the expectations hypothesis which is rejected but predict long-term interest rates and very significant. Our experts’research in the relevant theory is lagging behind. In the early1990s, the process of interest rates marketing on our country gradually began to constantly promote, on the process of interest rates marketing has brought a lot of influence on China’s economic development and financial market. In order to study and respond the influence which is brought by interest rate marketing, which we need to do an in-depth research the reasons for the term structure of interest rates and interest rate term structure itself for economic development and a lot of information in financial markets.As a result of using different mathematical models and selecting different data, domestic scholars got a different conclusion in the research of the expectations theory of the term structure of interest rates. Because of the difference between the models, it led to the using the same data come to different conclusions, this paper attempts to introduce the term premium model and compare it with the model does not introduce the term premium and conduct empirical analysis on the term premium is zero or constant or time-varying, using the SHIBOR data to test whether the interbank expectations theory of the term structure of interest rates is set up, and investigated whether the term premium is zero or constant, then study the influencing factors of the term premium by estimating the GARCH-M model.The innovations of this article:1From the perspective of long-term interest rate forecast, the empirical analysis was carried out on the expectations theory of term structure of interest rates, it compares with the same holding period for different maturities estimation results.2The Fama model introduces the term premium factor to predict the long-term interest rates and to make the parameter estimation. The model also estimates a constant term premium and time-varying term premium factors. Random error term may exist different variance and serial correlation which result in errors in the model. We use GMM method to estimate the model in order to avoid some bias. At the same time, we also compare the results of different estimation methods3We make out the term premium factor alone to take empirical analysis by estimating the GARCH-M model and the GARCH-M model which is given by Dziwura and Green to analyze the factors of the term premium factor.Through traditional linear regression model, we make a parameter estimation on the Fama model and the Fama model which introduces the revised term premium. The Fama model wthout considering time-varying term premium factor results in a "mystery expectations", which is same in the study of term structure of interest rates in United States by some foreign scholars. After the introduction of time-varying term premium factor, the test results of Fama model show that the rational expectation theory does not hold on, because premium factor in the yield curve is time-varying.Firstly, we use the GARCH-M model includes excess yield premium factor to fit the time period of volatility risk premium and get a better model finally. From the results of our estimate we can see the changing of term premium are influenced by income rate and excess return rate. With the increase of holding period, the income rate impact on term premiums is stronger, but the influence of the volatility is smaller. At last the results also show that yields and the volatility of excess return rate have a better forecast of term premiums. Last but not the least, make an estimate of GARCH-M model with the term premium factor and get the consistent conclusions.