The one "good" thing about this paper is that they don't even try to defend the arbitrariness of their model. Absolutely not "robust". EDIT: The copy I looked

The one "good" thing about this paper is that they don't even try to defend the arbitrariness of their model. Absolutely not "robust". EDIT: The copy I looked through was the published one. http://dx.doi.org/10.1016/j.ejor.2009.06.009 http://dx.doi.org/10.1016/j.ejor.2009.06.009