A multicriteria approach based on Analytic Hierarchy Process and compromise programming in portfolio selection

The traditional approach to the portfolio selection problem, the mean-variance model, shows limitations because it assumes that any investment opportunity could be meaningfully measured in terms of means and variances of returns. The model should include other criteria, based on liquidity, the characteristics of return distributions, or environmental factors. To incorporate in the model these additional relevant aspects of the problem, alternative methodologies are needed. This paper deals with the problem by using a multicriteria approach based on compromise programming and various options to incorporate the prior information (the norm and the criteria weights). A simplified version of the model has been applied to the stocks included in the Madrid Stock Market IBEX35 index. This article is dedicated to the memory of our colleague and friend Professor Rafael Santamaria, one of the most brilliant Spanish Professors of Finance, who passed away on June 18, 2018, at the age of 57.