Empirical research on time-varying characteristics and efficiency of the Chinese economy and monetary policy: evidence from the MI-TVP-VAR model

摘要

Based on the general time-varying parameter vector autoregressive model and data mining technology, this study proposes a new extension mixed innovation time-varying parameter stochastic volatility vector autoregressive model and investigates time-varying characteristics and efficiencies of different shock effects on China's monetary policy towards inflation and GDP. Using sample monthly data for 1979-2014, we utilize typical time points to illustrate the mechanisms between different economic variables via the Markov Chain Monte Carlo method and impulse response function. The empirical results show that the monetary transmission mechanism in China can be effective in the real economy, but with delay and efficiency leakage. The average delay and maximum efficiency can be measured through the MI model, which can capture accurate information of economic variables, effectively improving the precision of macroeconomic regulation and control. Meanwhile, the difference between the impacts of different channels is obvious; while the impact of interest rates is not significant, the impact of stock market is significant. The action mechanism between GDP and the inflation rate undergoes a gradual structural change, evidently displaying time-varying characteristics and a gradually weakening impact over time.