Textual Analysis in Real Estate

Abstract

Summary This paper incorporates text data from MLS listings into a hedonic pricing model. We show that the comments section of the MLS, which is populated by real estate agents who arguably have the most local market knowledge and know what homebuyers value, provides information that improves the performance of both in‐sample and out‐of‐sample pricing estimates. Text is found to decrease pricing error by more than 25%. Information from text is incorporated into a linear model using a tokenization approach. By doing so, the implicit prices for various words and phrases are estimated. The estimation focuses on simultaneous variable selection and estimation for linear models in the presence of a large number of variables using a penalized regression. The LASSO procedure and variants are shown to outperform least‐squares in out‐of‐sample testing. Copyright © 2016 John Wiley & Sons, Ltd.

Information Value of Property Description: A Machine Learning Approach

“…The evidence on the price impacts of frequent keywords in real estate advertisements thus far has been mixed. On one hand, Levitt and Syverson (2008), Rutherford and Yavas (2005), and Nowak and Smith (2017) find the inclusion of indicator variables for positive/negative words and short phrases in real estate advertisements can reduce omitted variable biases; On the other hand, Goodwin (2014) and Pryce (2008) point out the effects of positive/negative words on real estate prices are not consistent across different word classes.…”

Situating individuals’ subjective needs and aspirations within the wider context of the housing market

“…Similar to the approach used by Raamkumar (2016), the categories were predefined so that they could be compared with the automatically generated categories from the community algorithms. This is very different from studies such as that from Nowak and Smith (2017) terms. Thus, how a vertex was connected, shaped its identity.…”

Cultural Superstitions and Residential Real Estate Prices: Transaction-Level Evidence from the US Housing Market

“…Text has recently been used in real estate settings as well. Using a pre-specified dictionary of positive and negative words, Goodwin et al (2014) find the length and tone of written property descriptions significantly impact market outcomes, while Nowak and Smith (2016) identify which words in property descriptions are relevant when pricing real estate.…”