The Measure Of China Inflation Expectation And Inflation Forecasting

Since the reform and opening up, China’s market economic system has beengradually established and the level of productivity has been gradually increased.Along with the coming of the new period of economic development, inflation hasemerged. From1978to2012, China’s total GDP turned sixty times and per capitaGDP turned forty times. The average price of commercial housing was increased from703RMB per square meter in1990to5791RMB per square meter in2012. Theaccommodation and catering added value that was58.46billion RMB in1990wasincreased to1.04642trillion RMB in2012. With the sharp rise of price level, inflationand expectation become one focus of fervent discussions among economists. Thus,correct inflation forecasting has great significance to China’s economic development.The paper mainly studies China’s inflation expectation and the forecast ofinflation rate. Now the main research method to predict the inflation rate is based onNew Keynesian Phillips curve, so in this article we also study the inflation forecastingbased on the New Keynesian Phillips curve. What’s more, considering the importantinfluence of inflation expectation and the output gap when predict the inflation rate,we first measure inflation expectation and study the nonlinear relationship betweeninflation and the output gap. Then we will predict the inflation rate on the basis ofinflation expectation and output gap.First, we measure inflation expectation. Given the level that Chinese governmentcontrols the market economy as well as the use of deliberate fiscal policy, so choosinga reasonable target range of inflation trend has practical application value and the wayis in accordance with Chinese economic situation when measure the inflationexpectation. Because of the existence of the target range, the general state-spacemodel will lose efficacy. Instead we mainly use the approach of nonlinear state-spacemodel and the Metropolis-Hastings algorithm. The results show that the nonlinearstate-space model not only has more practical significance, but also to have bettermeasuring result for inflation expectation.Second, we use the nonlinear Granger causality test to study the nonlinear relationship between the inflation rate and the output gap. By the BDS statistics, theresult shows that the output gap has significant nonlinear structural feature so does theinflation rate. Then with the help of the nonparametric method we research thenonlinear Granger causality. The results show that there is an unidirectional nonlinearcausality relationship when the lag order is greater than7. The output gap nonlinearlyaffects the inflation rate, but the inflation rate does not affect the output gap.Therefore, when use econometric models to forecast the inflation rate, we can use thelinear model based on the output gap to predict the inflation rate.Last, we mainly adopt the semi-structural model that is based on the NewKeynesian Phillips curve and the characteristics of time series to predict the inflationrate. The paper selects hybrid New Keynesian Phillips under the open economy as thebenchmark model and introduces international crude oil supply shock and the realexchange rate of RMB against the dollar. What’s more, considering the importantinfluence of inflation expectation and the output gap, here we make use of theinflation expectation that is measured in the former chapter and construct asemi-structural four factors model including supply shock, demand shock, inflationinertia and inflation expectation to predict the inflation rate. Now GMM method isused to work on the relative estimation. By using the semi-structural model to predictthe inflation rate, the forecasting result shows that this kind of model has goodperformance when study future inflation forecasting and it can have themicroeconomic foundation of structural models and the dynamic feature of variablesof non-structural models.