On B-robust instrumental variable estimation of the linear model with panel data

Consistent estimation of linear panel data models with measurement error

“…This also defines a direction for further research, by bringing the literature on robust estimation in panel data models to bear on measurement error, cf. Wagenvoort and Waldmann (2002) and Bramati and Croux (2007).…”

Instrumental weighted variables under heteroscedasticity, Part I -- Consistency

“…But there was only a limited number of papers trying to robustify the method, see e. g. [1,9,10,11,34,35,36,38,39,40,50,73] or [57] (and the references given there 5 ). Moreover, even these papers, including [66], consider the situation when the disturbances are homoscedastic, except of [7] where the idea of the generalized method of moments (GMM) is employed, see [26].…”

Robust estimators for the fixed effects panel data model

“…Thus our aim is to build up an estimation procedure which is less sensitive to the presence of aberrant observations. In this first study we have focus on two different directions: the first one is based on a high breakdown Generalized M-estimator, similar to Wagenvoort and Waldmann (2002). A difference is that we use the LTS estimator (Rousseeuw, 1984) as initial regression estimator and a multivariate S-estimator to downweight leverage points.…”

“…A difference is that we use the LTS estimator (Rousseeuw, 1984) as initial regression estimator and a multivariate S-estimator to downweight leverage points. These estimators are more efficient and faster to compute than the ones used by Wagenvoort and Waldmann (2002). The second one is the MS estimator of Maronna and Yohai (2000) applied to the particular setting of the linear panel data model.…”