NotesFAQContact Us
Search Tips
ERIC Number: ED548790
Record Type: Non-Journal
Publication Date: 2012
Pages: 189
Abstractor: As Provided
Reference Count: N/A
ISBN: 978-1-2677-5210-9
Essays on Technology and Forecasting in Macroeconomics
Samuels, Jon Devin
ProQuest LLC, Ph.D. Dissertation, Johns Hopkins University
The three chapters in this dissertation use disaggregated models and data to provide new insights on well-established questions in macroeconomics. In the first chapter, to analyze how productivity impacts the business cycle, I model aggregate production with a production possibility frontier that accommodates sector-and factor-biased productivity. DSGE model simulations based on econometric estimates of the production frontier predict that in response to a labor-biased productivity shock, the macro economy expands; investment, consumption, and labor increase. In contrast, in response to a sector-biased technology shock, the macro economy exhibits contractionary-type responses; hours worked fall, and investment and consumption may fall as well. The confidence bands for the impact effects include both New Keynesian and RBC-type responses. Thus, it is difficult to distinguish between these classes of models based solely on the impact effect of a productivity shock. The second chapter uses the production possibility frontier to analyze the contribution of semiconductors to U.S. economic growth and productivity. I use a prototype NAICS-based industry production account to measure the direct impact of semiconductor production on aggregate growth and productivity, and the contribution of semiconductors via industries that use these devices as intermediate input. Using total factor productivity as a measure of innovation, I find that over the 1960-2007 period, innovation in the Semiconductor industry grew close to 9% per year, twenty five times the innovation growth rate for the economy as a whole, and semiconductors accounted for close to 30% of aggregate economic innovation. The third chapter, which is joint work with Rodrigo Sekkel, uses disaggregated data to investigate forecasting methods for macroeconomic aggregates. We argue that the simple average approach to combining predictions from a set of models can be improved by trimming the set of models prior to forecast combination. We compare trimming schemes and propose a new version based on the Model Confidence Set (Hansen et al. (2011)). We find significant gains in out-of-sample forecast accuracy from our proposed trimming method. We argue that persistence in forecasting performance and parameter estimation error in small samples provide an explanation for these gains. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page:]
ProQuest LLC. 789 East Eisenhower Parkway, P.O. Box 1346, Ann Arbor, MI 48106. Tel: 800-521-0600; Web site:
Publication Type: Dissertations/Theses - Doctoral Dissertations
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A