Search results

This is the first article to apply a machine learning approach to the analysis of time on market on real estate markets.

The random survival forest approach is introduced to the real estate market. The most important predictors of time on market are revealed and it is analyzed how the survival probability of residential rental apartments responds to these major characteristics.

Results show that price, living area, construction year, year of listing and the distances to the next hairdresser, bakery and city center have the greatest impact on the marketing time of residential apartments. The time on market for an apartment in Munich is lowest at a price of 750 € per month, an area of 60 m2, built in 1985 and is in a range of 200–400 meters from the important amenities.

The findings might be interesting for private and institutional investors to derive real estate investment decisions and implications for portfolio management strategies and ultimately to minimize cash-flow failure.

Although machine learning algorithms have been applied frequently on the real estate market for the analysis of prices, its application for examining time on market is completely novel. This is the first paper to apply a machine learning approach to survival analysis on the real estate market.

The COVID-19 pandemic, a sudden and disruptive external shock to the USA and global economy, profoundly affected various operations. Thus, it becomes imperative to investigate the repercussions of this pandemic on the US housing market. This study investigates the impact of the COVID-19 pandemic on a crucial facet of the real estate market: the Time on the Market (TOM). Therefore, this study aims to ascertain the net effect of this unprecedented event after controlling for economic influences and real estate market variations.

Monthly time series data were collected for the period of January 2010 through December 2022 for statistical analysis. Given the temporal nature of the data, we conducted the Durbin–Watson test on the OLS residuals to ascertain the presence of autocorrelation. Subsequently, we used the generalized regression model to mitigate any identified issues of autocorrelation. However, it is important to note that the response variable derived from count data (specifically, the median number of months), which may not conform to the normality assumption associated with standard regression models. To better accommodate this, we opted to use Poisson regression as an alternative approach. Additionally, recognizing the possibility of overdispersion in the count data, we also explored the application of the negative binomial model as a means to address this concern, if present.

This study’s findings offer an insightful perspective on the housing market’s resilience in the face of COVID-19 external shock, aligning with previous research outcomes. Although TOM showed a decrease of around 10 days with standard regression and 27% with Poisson regression during the COVID-19 pandemic, it is noteworthy that this reduction lacked statistical significance in both models. As such, the impact of COVID-19 on TOM, and consequently on the housing market, appears less dramatic than initially anticipated.

This research deepens our understanding of the complex lead–lag relationships between key factors, ultimately facilitating an early indication of housing price movements. It extends the existing literature by scrutinizing the impact of the COVID-19 pandemic on the TOM. From a pragmatic viewpoint, this research carries valuable implications for real estate professionals and policymakers. It equips them with the tools to assess the prevailing conditions of the real estate market and to prepare for potential shifts in market dynamics. Specifically, both investors and policymakers are urged to remain vigilant in monitoring changes in the inventory of houses for sale. This vigilant approach can serve as an early warning system for upcoming market changes, helping stakeholders make well-informed decisions.

Investors often utilize brokers to assist them in property acquisitions. These brokers are compensated through a cooperative commission, or bonus, that is publicized on the listing service. The purpose of this paper is to determine the relationship between advertised compensation packages and selling price, time-on-market and listing characteristics.

To examine variables likely to influence earnings of the buyers' broker, this study utilizes multiple and logistic regressions. Given the range of prices found in the 196,276 listings, the data was sorted on listing price and then split into ten, approximately equal, deciles.

The explanatory power of models with cooperative commission as the dependent variable was highest in the lowest deciles with type of financing, size and distressed status being highly significant. When comparing list- to selling price the average was 96.1%. As cooperative commission increased, the higher priced parcels sold at a higher price relative to list price. This potentially justifies higher cooperative commissions or exemplifies the principal-agent problem where effort is based on potential earnings. Fixed bonuses were used predominately for parcels under $62,234, likely to provide a minimum earnings amount. However, surrounding the median, it seems they may differentiate a property.

This research provides insight for practitioners on the impact of different variables, including cooperative commissions, on sale price and time-on-market. For example, cooperative commission increased for properties in the outer deciles implying that agents may be compensating for suspected difficulty. Additionally, the seasonality findings imply that agents can determine when to list and when to provide a fixed bonus to solicit attention. Results also suggest that practitioners will find it beneficial to market at an appropriate price rather than list high to create negotiating room.

This paper follows only one paper that covered a similar topic. However, this paper uses twenty years of multi-unit property listings from a major US city from 1996 to 2015. The focus on multi-unit properties is an effort to focus on a more sophisticated group of buyers that may be more experienced and make decisions more rationally.

This study aims to provide a framework for assessing the technical condition of a house to determine its market value, including the identification of other price-setting factors and their statistical significance. Time on market (TOM) in relation to the technical condition of a house is also addressed.

The primary database contains 631 houses, and the initial asking price and selling price are examined. All the houses are located in the Brno–venkov district in the Czech Republic. Regression analysis was used to test the influence of price-setting factors. The standard ordinary least squares estimator and the maximum likelihood estimator were used in the frame of generalized linear models.

Using envelope components of houses separately, such as the façade condition, windows, roof, condition of interior and year of construction, brings better results than using a single factor for the technical condition. TOM was found to be 67 days lower for houses intended for demolition – as compared to new houses – and 18 days lower for houses to refurbishment.

To the best of the authors’ knowledge, this paper is original in the substitution of specific price-setting factors for factors relating to the technical condition of houses as well as in proposing the framework for professionals in the Czech Republic.

This study aims to explore the spatial externalities of apartment vacancy rates on housing rent by considering multiple vacancy durations.

This research uses smart meter data to measure unobservable vacant houses. This study made a significant contribution by applying building-level smart meter data to housing market analysis. It examined whether vacancy duration significantly affected apartment rent and whether the relationship between apartment rent and vacancy rate differed depending on the level of housing rent.

The primary finding indicates that there is a significant negative correlation between apartment rent and vacancy duration. Considering the spatial externalities of apartment vacancy rates, the apartment vacancy rates of surrounding buildings did not show any statistical significance. Moreover, quantile regression results indicate that although the bottom 10% of apartment rent levels showed a negative correlation with all vacancy durations, the top 10% showed no statistical significance related to vacancies.

This study measures the extent of spatial externalities that can differentiate taxation based on housing vacancies.

The findings indicate that landlords have asymmetric information about their buildings compared with the surrounding buildings, and the extent to which price adjusts for long-term vacancies differs depending on the level of apartment rent.

With access to the large-scale search data from Rightmove plc, the paper firstly indicated the possibility of using user-generated data from online property portals to predict housing market activities and secondly embraced a GIS approach to explore what people search for housing and what they chose and investigated the issue of mismatch between search patterns and revealed patterns. Based on the analysis, the paper contributes a visual GIS-based approach which may help planners and designers to make more informed decisions related to new housing supply, particularly where to build, what to build and how many to build.

The paper used the 2013 housing search data from Rightmove and the 2013 price data from Land Registry with transactions made after the search period and embraced a GIS approach to explore the potential housing demand patterns and the mismatch between searches and sales. In the analysis, the paper employed the K-means approach to group prices into five levels and used GIS software to draw maps based on these price levels. The paper also employed a simple analysis of linear regression based on the coefficient of determination to investigate the relationship between online property views and values of house sales.

The result indicated the strong relationship between online property views and the values of house sales, implying the possibility of using search data from online property portals to predict housing market activities. It then explore the spatial housing demand patterns based on searches and showed a mismatch between the spatial patterns of housing search and actual moves across submarkets. The findings may not be very surprising but the main objective of the paper is to open up a potentially useful methodological approach which could be extended in future research.

It is important to identify search patterns from people who search with the intention to buy houses and from people who search with no intention to purchase properties. Rightmove data do not adequately represent housing search activity, and therefore more attention should be paid to this issue. The analysis of housing search helps us have a better understanding of households' preferences to better estimate housing demand and develop search-based prediction models. It also helps us identify spatial and structural submarkets and examine the mismatches between current housing stock and housing demand in submarkets.

The GIS approach in this paper may help planners and designers better allocate land resources for new housing supply based on households' spatial and structural preferences by identifying high and low demand areas with high searches relative to low housing stocks. Furthermore, the analysis of housing search patterns helps identify areas with latent demand, and when combined with the analysis of transaction patterns, it is possible to realise the areas with a lack of housing supply relative to excess demand or a lack of latent demand relative to the housing stock.

The paper proves the usefulness of a GIS approach to investigate households' preferences and aspirations through search data from online property portals. The contribution of the paper is the visual GIS-based approach, and based on this approach the paper fills the international knowledge gap in exploring effective approaches to analysing user-generated search data and market outcome data in combination.

This theoretical study aims to clarify the (a priori) ambiguous effect of homeownership on unemployment. In general, in fact, homeownership discourages job mobility, but homeowners are less likely to be unemployed than tenants, since homeownership would seem to be positively related to human capital.

This study develops a modified version of the benchmark theoretic model of the labour market – the well-known “equilibrium unemployment theory” – where homeownership affects both the “Beveridge Curve” (BC, by means of job immobility) and the “Job Creation Condition” (JCC, by means of human capital).

The general result is that an increase in homeownership increases unemployment. Therefore, policymakers could encourage job mobility, before facilitating homeownership. This policy implication, however, may not apply in the case of high inflation and/or low nominal interest rate, and when the job destruction rate depends on the homeownership rate.

The model studies the steady-state equilibrium of the labour market, so the policy implications only relate to the long-run. The model, therefore, does not consider the short-run effects of homeownership on unemployment (which may differ from the long-term results).

The model suggests a public policy characterised by large investment in rail lines and subsidised commuter fares. By promoting a more efficient allocation of workers across regions (and, thus, job mobility), indeed, this policy can be a good way to increase employment, without harming homeownership.

The practical implication of this model is also a social implication, since it relates to homeownership and housing tenure.

To the best of author’s knowledge, this is the first model that introduces the key role of homeownership in the so-called “Equilibrium unemployment theory”. Precisely, the model uses a modified version of both the BC (which includes the negative effect of homeownership on the overall job search intensity of unemployed workers) and the JCC (which includes the positive effect of homeownership on both the business start-up and the human capital of workers). By comparing these two opposite effects, this theoretical work makes clearer the net effect of homeownership on unemployment.