论文部分内容阅读
传统的Granger因果检验方法受阻于非平稳数据与时变因果关系的存在,本文以无限状态的Markov过程为基础,设计混合分层结构Gibbs算法,实现区制时变VAR模型,用于非平稳数据的时变因果关系检验。实证分析我国经济环境中,需求拉动效应的时变特征。结果表明:国内需求逐渐成为更加重要的经济增长动力;经济增长降低了对投资和出口的依赖;在增长方式的转变过程中,内需效应易受外需变动的影响而波动,需要适时的货币政策调节和投资效应的推动以保证稳定的经济增长,以利于进一步升级产业结构。
The traditional method of Granger causality test is hindered by the existence of non-stationary data and time-varying causality. Based on infinite Markov process, a hybrid hierarchical Gibbs algorithm is designed to implement the time-varying VAR model in the area, which is applied to non-stationary data Time-varying causal relationship test. Empirical analysis of China’s economic environment, the demand-pull effect of time-varying characteristics. The results show that domestic demand is gradually becoming a more important driving force for economic growth. Economic growth reduces the dependence on investment and exports. During the transformation of growth mode, the domestic demand effect is subject to fluctuations due to changes in external demand. Therefore, timely adjustment of monetary policy is needed And investment promotion to ensure stable economic growth in order to facilitate the further upgrading of industrial structure.