Identifying Real Estate Opportunities Using Machine Learning

Abstract

The real estate market is exposed to many fluctuations in prices because of existing correlations with many variables, some of which cannot be controlled or might even be unknown. Housing prices can increase rapidly (or in some cases, also drop very fast), yet the numerous listings available online where houses are sold or rented are not likely to be updated that often. In some cases, individuals interested in selling a house (or apartment) might include it in some online listing, and forget about updating the price. In other cases, some individuals might be interested in deliberately setting a price below the market price in order to sell the home faster, for various reasons. In this paper, we aim at developing a machine learning application that identifies opportunities in the real estate market in real time, i.e., houses that are listed with a price substantially below the market price. This program can be useful for investors interested in the housing market. We have focused in a use case considering real estate assets located in the Salamanca district in Madrid (Spain) and listed in the most relevant Spanish online site for home sales and rentals. The application is formally implemented as a regression problem that tries to estimate the market price of a house given features retrieved from public online listings. For building this application, we have performed a feature engineering stage in order to discover relevant features that allows for attaining a high predictive performance. Several machine learning algorithms have been tested, including regression trees, k-nearest neighbors, support vector machines and neural networks, identifying advantages and handicaps of each of them.

Predicting property prices with machine learning algorithms

“…It is also increasingly popular to apply machine learning techniques in the appraisal and prediction of property values. The merit of this technique is that it allows more recent property data or transactions to possibly 'correct' the parameters of existing model estimations, which were based on earlier observations (Baldominos et al, 2018;Hastie et al, 2004;Hausler et al, 2018;Rafiei & Adeli, 2016;Rogers & Girolami, 2011;Sun et al, 2015).…”

Modeling of apartment prices in a Colombian context from a machine learning approach with stable-important attributes

“…We used the random forest algorithm (formula, data, mtry, ntree, etc.) with the following parameters: mtry at its default value (p = m / 3; the subset size for the predictors); based on Baldominos et al [7] we would use ntree = c (10, 20 and 50), but only 10 trees were sufficient to achieve high prediction capacity.…”

“…marginal effects) [3,4]. These considerations must be taken into account when facing practical challenges such as: (i) the need to take advantage of new sources of massive data in real estate, in order to favor better decision making [6]; (ii) building models of real estate prices based on machine learning, with high predictability in non-training samples, and whose attributes are stables (considering different samples) [3,[7][8][9]; (iii) exploring patterns derived from texts as attributes of these models [7,10]; and (iv) using precise criteria when establishing policies related to real estate prices, thus preventing arbitrary fixations that are not supported by evidence [1,11].…”

Testing the Effectiveness of Outlier Detecting Methods in Property Classification

“…Authors proved that their research had successfully proposed an investment tool that rapidly identifies outstanding REIT stocks using two outlier detection algorithms. Detection of market opportunities has also been of interest in the Spanish real estate market (Baldominos, et al, 2018). The authors explored the application of diverse machine learning techniques with the objective of identifying real estate opportunities for investment.…”