Book chapter
Unbalanced Panel Data Models with Interactive Effects
The Oxford Handbook of Panel Data, pp.149-170
Oxford University Press
2015
DOI: 10.1093/oxfordhb/9780199940042.013.0005
Abstract
Unbalanced panel data are common in empirical research. This chapter provides two types of estimators for panel data models in the presence of interactive effects and missing observations. One deals with the case when the common factors are deterministic and smooth in the time domain, and the proposed estimator is based on an iterative functional principal components analysis. The other method integrates the EM algorithm and the traditional principal components method, which effectively estimates the model when the factors are non-smooth and probably stochastic. Both static and dynamic panel models are considered. Extensive simulations are carried out. It is found that the EM-based methods are in general consistent whether the factors are smooth or not for both static and dynamic models.
Details
- Title: Subtitle
- Unbalanced Panel Data Models with Interactive Effects
- Creators
- Jushan Bai - Columbia UniversityYuan Liao - University of Maryland, College ParkJisheng Yang - Huazhong University of Science and Technology
- Contributors
- Badi H Baltagi (Editor) - Syracuse University
- Resource Type
- Book chapter
- Publication Details
- The Oxford Handbook of Panel Data, pp.149-170
- DOI
- 10.1093/oxfordhb/9780199940042.013.0005
- Publisher
- Oxford University Press; New York, NY
- Language
- English
- Date published
- 2015
- Academic Unit
- Economics
- Record Identifier
- 9984937788302771
Metrics
1 Record Views