Measure of firms inefficiency and heterogeneity using Nonparametric frontier Models: The Côte d’Ivoire Case

The objective of this work is to estimate the production frontier and technical efficiency levels of Côte d’Ivoire firms. The methodology used is closely linked to the recent nonparametric frontier models, as for example Cazals, Florens and Simar (2002), Aragon, Daouia and Thomas-Agnan (2003) or Florens and Simar (2001). The main advantage compared to the classical nonparametric methods (DEA, FDH) is the robustness to outliers and we show that these latest technics are in particular more adapted to our database. We also modelize the technical efficiency by introducing external environment factors that might influence the production process. Using bootstrap technics, and following the ideas from Simar and Wilson (2003), we take into account the heterogeneity of our database and explain the variability of firms performance.