This thesis discusses the difficulties of pricing real estate by using the existing financial economic theory and investigates an alternative approach in order to take into account the 'unique' nature of real estate risk. It consists of three essays placing great emphasis on the risk handling problem. The first essay examines the risk-return trade-off of a mixed-asset portfolio that includes real estate using copula functions to model the dependence between real estate and financial assets. In particular, it analyses the role of direct as opposed to securitised real estate (mutual funds) in terms of diversification when the dependence structure is modelled by an appropriate copula. The empirical analysis is conducted using Swiss data for the period 1988-2008. It is shown that a better portfolio diversification is obtained with indirect rather than with direct real estate. The second essay addresses methodological issues in real estate pricing. Special attention is devoted to outlining pitfalls of the traditional income-based approach in terms of risk consideration. The evidence against the risk-adjusted discount rate's consistency with the cash flow's risk leads to an alternative approach based on the certainty equivalent cash flow for computing reservation prices. We include a precise definition of illiquidity as a measurable concept, formalize the computation of the illiquidity premium as well as the optimal marketing strategy for the seller. The third essay employs a detailed analysis of the price formation process to argue for the linkage of probable-price determination and the time necessary for a property sale. Within this framework, the factors accounting for the time required for a property sale are identified and used to explain the contradictions in the statistical studies found in the literature.
EPFL2009
