Inference on the Correlation between Permanent and Transitory Shocks for Unidentified Unobserved Components Models
20 Pages Posted: 23 Apr 2007 Last revised: 4 Jul 2008
Date Written: July 1, 2007
Abstract
In this paper, I propose a simple methodology for inferring the correlation between permanent and transitory shocks in unidentified unobserved components (UC) models, where the correlation is not identified. However, I show that there is an upper bound of the correlation implied from the unrestricted ARIMA reduced form. I apply the proposed methodology to GDP data of six G7 countries. It is found that the implied upper bounds of the correlation range from -0:75 to 0:99 depending on countries. This means that the two shocks are highly negatively correlated for these G7 countries. The paper also shows that, in the estimation of UC models, imposing a proper identification restriction is important for the estimation of trend and cycle by the estimated UC models.
Keywords: Unobserved components model, Trend, Cycle, Business Cycle, Correlated Components
JEL Classification: C13, C22, E32
Suggested Citation: Suggested Citation
0 References
0 Citations
Do you have a job opening that you would like to promote on SSRN?
- Citations
- Citation Indexes: 1
- Usage
- Abstract Views: 918
- Downloads: 115
- Citations
- Citation Indexes: 1
- Usage
- Abstract Views: 918
- Downloads: 115