Recursive solution methods for dynamic linear rational expectations models (Q911206): Difference between revisions

The author presents a new estimation technique for dynamic linear rational expectations models by use of the state-space representation. Compared with alternative estimation procedures this approach has several advantages: first, the constraints that the model places on the data are transparent; second, the model can be solved recursively by the Kalman filter, thus, avoiding algebraic model solutions; and third, modifications in the empirical model as well as in the likelihood function are easily incorporated. Prior to the presentation of the general model the author clarifies his approach using a simple two variable model. In section 4 it is shown how to estimate the model. Finally, the author gives an example by studying the relationship between stock prices and dividends.