Data uncertainty: Empirical evidence, generalequilibrium implications, and hedging strategies
In the first chapter we document the empirical properties of revisions to major macroeconomic variables in the U.S., over the period 19662000. We find that these revisions do not have a zero mean, which indicates that the initial announcements by statistical agencies are biased. We also find that the revisions are quite large compared to the original variables. They are predictable using the information set at the time of the initial announcement, which means that the initial announcements of statistical agencies are not rational forecasts. We also provide some evidence that professional forecasters ignore this predictability. Our findings suggest that data revisions in the U.S. do not satisfy simple desirable statistical properties. In the second chapter, using the empirical results from the previous chapter as the motivation, we study the effects of data revisions in a general equilibrium framework. We find that the presence of data revisions, or data uncertainty, creates a precautionary motive and causes significant changes in the decisions of agents. We also find that the model with revisions captures some aspects of the business cycle dynamics of the US data better than the benchmark model with no revisions. Using our model we measure the cost of having data revisions to be about $33 billion, $5 billion of which can be recovered by eliminating the predictability of revisions. Comparing these numbers with the budgets of the major statistical agencies in the US, we conclude that any money spent on the improvement of data collection would be well worth it. In the third chapter we take the first step in analyzing ways of hedging the risk of data revisions. We show that we can construct portfolios of assets, which are maximally correlated with revisions to macroeconomic variables. The average correlation of the returns of these hedging portfolios and the underlying revision risk are in the order of 0.50, which is very promising in terms of hedging performance.
Year of publication: 
20040101


Authors:  Aruoba, S. Boragan 
Publisher: 
ScholarlyCommons 
Saved in favorites
Similar items by person

Comparing solution methods for dynamic equilibrium economies
Aruoba, S. Boragan, (2003)

The Macroeconomy and the Yield Curve: A Nonstructural Analysis
Francis X. Diebold, (2003)

Moving Holidays and Seasonality: An Application in the Time and Frequency Domains For Turkey
Alper, C. Emre, (2001)
 More ...