Increasing the Value of Search Subscriptions for Housing Market Analyses
Posted: 1 Nov 2017 Last revised: 19 Dec 2018
Date Written: October 30, 2017
The drastic increase in the number of vacant accommodations in some regions of Switzerland and the simultaneous housing shortage in others are the result of not knowing where people want to live and, therefore, of having built accommodations in the wrong locations. In order to better understand what people are searching for, the Swiss start-up Realmatch360 began to analyze search subscriptions to real estate platforms. Using search subscriptions allows the company to get a better understanding of people’s preferences for housing and even to identify unmet demand. In this paper, we propose powerful approaches based on unsupervised learning to maximize the benefits of using search subscriptions exhibiting many missing entries for housing market analyses.
Keywords: Demand Forecasting, Imputation, Machine Learning, Real Estate
Suggested Citation: Suggested Citation