Abstract
This paper studies the estimation of panel cointegration models with cross-sectional dependence generated by unobserved global stochastic trends. We extend the CupFM (continuously updated and fully modified) estimators studied by Bai et al. (2009) by allowing cross-sectional heterogeneity. To remove the endogeneity caused by the cointegrated triangular system, we consider the Continuously Updated Dynamic OLS (CupDOLS). The consistent limiting distribution of the estimators is derived.
| Original language | American English |
|---|---|
| State | Published - Jan 1 2012 |
| Event | International Chinese Statistical Association Applied Statistical Symposium - Boston, United States Duration: Jun 23 2012 → Jun 26 2012 |
Conference
| Conference | International Chinese Statistical Association Applied Statistical Symposium |
|---|---|
| Abbreviated title | ICSA |
| Country/Territory | United States |
| City | Boston |
| Period | 06/23/12 → 06/26/12 |
Disciplines
- Business Administration, Management, and Operations
- Operations and Supply Chain Management
Keywords
- Cross-sectional dependence
- Cross-sectional heterogeneity
- Dynamic OLS estimator
- Factor analysis
- Fully-modified estimator
- Panel cointegration