Inferring Changes in Expectation Behavior Over Time: An Application of Nonlinear Time-Varying-Parameters Estimation
提出一种新方法，从宏观经济时间序列中推断代理人预期结构的变化，通过非线性时变参数估计货币增长模型的权重，揭示代理人如何随时间修正对竞争模型的信念。
Abstract This article develops a new method for inferring the structure of agents' expectations from macroeconomic time series. Two versions of a somewhat stylized macroeconomic model are examined. Expectations of money growth are a central driving variable, assumed to be formed as a time-varying weighted average of alternative money-growth models. The weights are estimated as time-varying parameters jointly with the structural parameters for the model, using a nonlinear time-varying-parameters method developed for the article. The results allow us to infer how agents may have revised their beliefs about competing money-growth models (including Federal Open Market Committee announcements) over the sample. KEY WORDS: Extended Kalman filterRational expectations