This project develops a model which can better achieve an overall integration between real and financial phenomena than the representative agent neoclassical growth model used in real business cycle analysis. This would be an important contribution because the models employed in Real Business Cycle theory appear to be unable to account for the volatility of asset prices at business cycle frequency or the spread between the average annual returns of Treasury Bills and the return on equities. The model developed in this project differs both in its impulse (recurring uncertainty) and propagation mechanisms (the choice between flexible and inflexible actions) from Real Business Cycle theory. Shocks to the economy are only partially observable. As a result, the representative agent draws an inference about the true shock from those which are observable. These inferences are subject to recurring fluctuations in "confidence." By calibrating its parameters to long-run growth facts of the U.S. economy, key statistics of the actual economy are reproduced through the model economy. More specifically, the model reproduces real magnitudes such as consumption, output and investment, and observed financial statistics on the volatility of asset prices, the premium on equities, and the term structure of interest rates.
