Learning outcomes

In this chapter, you will learn how to

• undertake all the stages involved in designing, building and evaluating an empirical econometric model in real estate through two detailed examples.

The regression analysis topics of chapters 4 to 6 are fundamental to conducting empirical research in real estate. Given the importance of regression in real estate analysis, we devote this chapter to more examples of multiple regression. We give two detailed illustrations, with the aim of further familiarising the reader with key tasks in the empirical estimation of a multiple regression model and particularly in model construction and selection. In chapters 4 to 6, we examined regressions with time series data, the most common form of data in real estate modelling. This chapter also presents an example with cross-sectional data. The estimation principles are identical but the focus of the illustration is also on the interpretation and usage of cross-section analysis.

Frankfurt office rents: constructing a multiple regression model

We focus on modelling office rents using a multiple regression model and use data for the Frankfurt office market. Our specification is guided by theory, as any good empirical model of the real estate market should be. Hence the empirical model of Frankfurt office rents in this example is driven by the a priori treatment of office rent determination originating in theoretical frameworks of real estate market dynamics put forward in a number of studies (e.g. DiPasquale and Wheaton, 1992; Clapp, 1993; RICS [Royal Institution of Chartered Surveyors], 1994; Ball, Lizieri and MacGregor, 1998).