Guimaraes, Paulo Figueiredo, Octávio Woodward, Douglas
Year of Publication:
42nd Congress of the European Regional Science Association: "From Industry to Advanced Services - Perspectives of European Metropolitan Regions", August 27th - 31st, 2002, Dortmund, Germany
Given its sound theoretical underpinnings, the Random Utility Maximization-based conditional logit model has been the methodological basis for applied research on industrial location decisions. However, in practice, the implementation of this methodology presents problems. A notable one is the underlying Independence of Irrelevant Alternatives (IIA) assumption. In this paper we show that by taking advantage of an equivalence relation between the likelihood function of the conditional logit model and the Poisson regression [Guimarães, Figueiredo and Woodward (2002)] one can more effectively control for the potential IIA violation resulting from omitted attribute characteristics. We also provide an empirical illustration, wherein we exemplify how that relation can be helpful to investigate the location determinants of new manufacturing plants in the United States counties.