Abstract
This paper studies what happens when we move from a short regression to a long regression in a setting where both regressions are subject to misspecification. In this setup, the least-squares estimator in the long regression may have larger inconsistency than the least-squares estimator in the short regression. We provide a simple interpretation for the comparison of the inconsistencies and study under which conditions the additional regressors in the long regression represent a “balanced addition” to the short regression.
| Original language | English |
|---|---|
| Pages (from-to) | 1183-1200 |
| Number of pages | 18 |
| Journal | Journal of Economic Surveys |
| Volume | 32 |
| Issue number | 4 |
| Early online date | 13 Feb 2018 |
| DOIs | |
| Publication status | Published - Sept 2018 |
Funding
The authors are grateful to Eveline de Jong for providing the example in the introduction; and to Ed Leamer, seminar participants at Tilburg University (May 2016), Georgetown University (October 2016), ICEEE (January 2017), and EEA-ESEM (August 2017), the editor and two referees for constructive comments and useful suggestions. Giuseppe De Luca and Franco Peracchi also acknowledge financial support from MIUR PRIN 2015FMRE5X.
| Funders | Funder number |
|---|---|
| EEA-ESEM | |
| ICEEE | |
| Georgetown University | |
| Ministero dell’Istruzione, dell’Università e della Ricerca | |
| Universiteit van Tilburg |
Keywords
- Bias amplification
- Inconsistency
- Least-squares estimators
- Mean squared error
- Omitted variables
- Proxy variables