Robust models for style analysis

This study contributes to the literature on style analysis by extending the quantile-based style model proposed by Basset and Chen (2001) through the introduction of the operational limits the manager faces up in handling an investment portfolio. In this framework, a practical guidance concerning the implementation of the quantile-based style model, the efficient use the obtained results, the computational complexity due to the introduction in the model of linear inequality constraints is presented. Validation issues of the proposed model are also discussed and more flexible solution for inference purposes based on resampling schemes is suggested. The efficiency and the robustness of the proposed alternative approach is explored through a simulation study.