A Statistical Model with Qualitative Input

Abstract

A statistical estimation model with qualitative input provides a mechanism to fuse human intuition in the form of qualitative information into a statistical model. We investigate the statistical properties of this model and devise a numerical computation method for a model subclass with a uniform correlation structure. We show that, within this subclass, qualitative information can be as useful as quantitative information. We also show that the correlation between variables compromises the accuracy of the statistical estimate. However, the adverse effect from the correlation can be minimal, as is illustrated in an application to portfolio selection. The proposed model, when used in conjunction with approximation techniques, is shown to have potential for portfolio selection with financial data.