Inflation Expectation: Heterogeneity And Dynamic Process

This paper focuses on the heterogeneity and dynamic updating of inflationexpectations. We divide people into two groups, resident and expert, to comparethe inflation expectations between them. We find that there is the heterogeneitybetween the two groups, and the inflation expectation of expert is more closed toactual inflation than resident’s.Then we turn to the inflation updating process of resident. From a prospectiveof signal extraction, we build a Bayesian state space model to connect theinflation expectation of resident and expert. For the first time in China, we try toincorporate the influence of media report into the formation of inflationexpectations. To estimate the model, we use Gibbs sampling algorithm and findsome interesting result:(1) the inflation expectation of the resident will adjustwith the time;(2) media report can influence the inflation expectation of theresident;(3) the more the media report, thus the more the information, the moreaccurate the inflation expectation of the resident.