Data Analysis and Applications 1

9An SEM Approach to Modeling Housing Values

Although hedonic regression remains a popular technique for estimating property values, structural equation modeling (SEM) is increasingly seen as a realistic analytical alternative. This chapter presents an SEM analysis of a historical data set for a large Canadian realtor. An iterative approach was adopted for the modeling. The first phase focused on internal relationships between houses’ structural characteristics and the second on housing values and their determinants. In the final phase, advertised list prices and location details were the priority. A comprehensive evaluation of the resulting holistic model revealed a wealth of significant structural relationships – particularly between house style, structure and attributes.

9.1. Introduction

Housing is a durable, highly differentiated asset, characterized by being fixed in location (Kinnard 1968). Effective price estimation (Karanka et al. 2013) is crucial to its successful acquisition. With Hedonic Price Theory (Rosen 1974), a house is considered a “basket” of attributes – z₁, z₂, …, z_n – against which the house price, P = f(z₁, z₂, …, z_n), can be derived. Here, attributes are typically related to structural, locational, neighborhood and environmental (e.g. noise and pollution) characteristics (Anas and Eum 1984). Depending on the functional form f adopted for computing P, various hedonic regression formulations are available for estimation purposes (Palmquist 1984). Some ...