Fixed effects, random effects or Hausman–Taylor?: A pretest estimator

366 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Big or Small Cities? On city size and economic growth (2017, positive): “The HT has the advantage over a fixed effects estimator that it allows for the inclusion of time-invariant variables in a panel setting, e.g. land area as well regional and oil dummies in our case (Baltagi, Bresson, and Pirotte 2003).”
Does the Nonfarm Economy Offer Pathways for Upward Mobility? Evidence from a Panel Data Study in Ethiopia (2012, methodology): “…the coefficients for both the time-varying and timeinvariant variables consistently, this method use the time-varying variables that are uncorrelated with the latent variable to instrument the variables correlated with the latent variable i (see Hausman and Taylor, 1981; Baltagi et al., 2003).”
On the Fixed-Effects Vector Decomposition (2011, background): “One conventional approach to finding a balance would be to select between the competing estimators based on a specification test (Baltagi et al., 2003). If the test rejects the null hypothesis of no difference between estimators, then ht would be selected. Otherwise, the efficient estimator estimato”
The impact of a Hausman pretest on the size of a hypothesis test: The panel data case (2010, background): “See Baltagi et al. (2003) for the finite sample performance of a Hausman and Taylor (1981) type pretest estimator.”