Real Estate in the Real World: Dealing with Non-Normality and Risk in an Asset Allocation Model

Abstract

Executive Summary. Quantitative models of asset allocation are increasingly used by institutional commercial real estate investors as a guide for investment strategy. Real estate as an asset class, however, does not conform well to many of the assumptions underlying standard mean-variance optimization. This paper outlines a model of allocation that addresses two important “real world” violations of these assumptions. First, the assumption that returns are normally distributed is relaxed; instead, returns are modeled using a distribution that allows for both the “fat-tailed” behavior and skewness seen in asset returns. Second, an alternative to the traditional MPT optimizer is employed—the so called “downside deviation” model—that better reflects the observed behavior of investors.