|
Abstract:
An economic agent who is uncertain of her economic model learns, and this learning is sensitive to the presence of data measurement error. I investigate this idea in an existing framework that describes the Federal Reserve's role in U.S. inflation. This framework successfully fits the observed inflation to optimal policy, but fails to motivate the optimal policy by the perceived Philips curve trade-off between inflation and unemployment. I modify the framework to account for data uncertainty calibrated to the actual size of data revisions. The modified framework ameliorates the existing problems by adding sluggishness to the Federal Reserve's learning: the key point is that the data uncertainty is amplified by the nonlinearity induced by learning. Consequently there is an explanation for the rise and fall in inflation: the concurrent rise and fall in the perceived Philips curve trade-off.
Full paper, latest version (287 KB PDF)
| Full paper (screen reader version)
Original version (563 KB PDF)
Keywords
Data uncertainty, data revisions, real time data, optimal control, parameter uncertainty, learning, extended Kalman filter, Markov-chain Monte Carlo
PDF files: Adobe Acrobat Reader ZIP files: PKWARE
Home | IFDPs | List of 2008 IFDPs
Accessibility |
Contact Us
Last update: October 26, 2009
|
