Abstract
The weighted-average least squares (WALS) approach, introduced by Magnus et al. (2010) in the context of Gaussian linear models, has been shown to enjoy important advantages over other strictly Bayesian and strictly frequentist model-averaging estimators when accounting for problems of uncertainty in the choice of the regressors. In this paper we extend the WALS approach to deal with uncertainty about the specification of the linear predictor in the wider class of generalized linear models (GLMs). We study the large-sample properties of the WALS estimator for GLMs under a local misspecification framework, and the finite-sample properties of this estimator by a Monte Carlo experiment the design of which is based on a real empirical analysis of attrition in the first two waves of the Survey of Health, Aging and Retirement in Europe (SHARE).
| Original language | English |
|---|---|
| Pages (from-to) | 1-17 |
| Number of pages | 17 |
| Journal | Journal of Econometrics |
| Volume | 204 |
| Issue number | 1 |
| Early online date | 12 Jan 2018 |
| DOIs | |
| Publication status | Published - May 2018 |
Funding
We thank Luigi Augugliaro, Gerda Claeskens and Henk Pijls for discussions, Xinyu Zhang for his MATLAB code, and an Associate Editor and three anonymous referees for helpful comments. Giuseppe De Luca and Franco Peracchi acknowledge financial support from the MIUR PRIN 2015FMRE5X.
| Funders | Funder number |
|---|---|
| Ministero dell’Istruzione, dell’Università e della Ricerca | PRIN 2015FMRE5X |
Keywords
- Attrition
- Generalized linear models
- Model averaging
- Monte Carlo
- WALS