The Nonlinear Model Of Forward-Looking Taylor Rules And Test

In recent years, the monetary policy rule caught people’s eyes widely. Especially McCallum proposed McCallum rule and Taylor proposed the Taylor rule and after they putting the specific formulation of rule-based monetary policy into practice, it aroused great interest of an increasing number of scholars and the monetary authorities of all countries. And people gradually shifted their attention from the discretionary choices and the debate of different monetary policy rules to the discussion of the monetary authorities’ operation practices. Although monetary rule restrained the monetary authorities’ policy operation, the restraint impelled monetary policy to set up a credible commitment device. Therefore, the short-sighted and opportunism of monetary policy can be avoided. China’s monetary policy has always been in a strong discretionary choice policy, especially before the 21 st century, the government tightened macro-control and the economy is volatile. But in recent years, many studies showed that tools of monetary policy rules have some applicability in China, especially the Taylor rule which domestic scholars did much research, however, the most of the studies were limited in the traditional linear research of Taylor rule. And in the existing nonlinear research of Taylor rule, the related scholar ignored the endogenous structure which may exist in the model, and the neglect of the endogenous will inevitably lead to errors in the estimation of the model parameters.Based on this, this paper in the linear forward Taylor rule model based on focus on the inside nature structure of variable parameters of the Taylor rule and the quantile regression model, the concrete form of the model is given, two step method to estimate the parameters in the model theory, in order to test the nonlinear characteristics of the monetary policy of our country. Inspection and the monetary authorities in the different historical periods of the dynamic evolution path, through the monetary policy rule of the quantile regression can test different rate scales, monetary authorities efforts to regulate and control, can be through the analysis of monetary policy rules when degeneration.Considering the influence of monetary authorities to adjust the interest rate smoothing preference and Expectation Factors of monetary policy. Firstly, this paper based on interest rate smoothing type forward-looking Taylor rule by GMM method, the 1996 first quarter to the fourth quarter of 2014 monetary policy. The empirical results show that the model can better depict China’s monetary authorities in the regulation of behavior. Subsequently, in order to test the control preferences of different historical periods in China’s monetary authorities, is introduced on the basis of interest rate smoothing type forward-looking Taylor rule variable parameters, constructed in the time variable parameters of the Taylor rule, and based on the state space Kalman filter of two step estimation research. Then, in order to test the level of different interest rates, the monetary authorities’ regulation and control, this paper, based on the interest rate smoothing of the forward-looking Taylor rules, the use of endogenous quantile regression estimation. The above results show that the interest rate operation in our country has a very strong smoothing mechanism, the interest rate adjustment will be based on the last period of interest rate adjustment. China’s monetary policy response to inflation is inadequate, the steady price, China’s monetary policy is an unstable control mechanism, but the monetary authorities to inflation reaction coefficient showed upward trend. Monetary authorities of the output gap of the reaction coefficient is greater than 0, which shows that the steady growth of China’s monetary policy is stable, and the monetary authorities to the output gap is also constantly strengthened regulation and control. Finally, compared with the linear Taylor rule model and the time varying parameter model, the prediction accuracy of the Taylor rule is obtained, which is higher than that of the linear Taylor rule.