Aplicaciones del Filtro de Kalman a las calibraciones en modelos de ciclo real

Laburpena

This paper pursues two objectives. One is to generalize the Kalman Filter to dynamic models with rational expectations which include current expectations of future endogenous variables. A second objective is to illustrate two applications of this estimation procedure to stochastic rational expectations growth models. The first applications is a propasal to calibrate sorue parameters in tbese models whose estimation is difficult because of the lack of appropriate data (for example, the coefficient of relative risk aversion). In the second application, the previous calibration procedure is used to offer an objective measure which allows for discriminating among alternative models that have, in some aspects, a similar stochastic behaviour.