Search results

Planning for future real estate and facility needs in a highly uncertain competitive environment can benefit from a four‐stage process of demand forecasting. Based upon research conducted within the Corporate Real Estate Portfolio Alliance and a review of general business forecasting techniques, each successive stage relies on more abstract data and increased dialogue about business strategy with the lines of business as uncertainty about the future increases. Each stage requires increasing flexibility in the supply of real estate as the range of probabilities around the forecast widens.

The purpose of this paper is to examine internet search query data provided by “Google Trends”, with respect to its ability to serve as a sentiment indicator and improve commercial real estate forecasting models for transactions and price indices.

This paper examines internet search query data provided by “Google Trends”, with respect to its ability to serve as a sentiment indicator and improve commercial real estate forecasting models for transactions and price indices.

The empirical results show that all models augmented with Google data, combining both macro and search data, significantly outperform baseline models which abandon internet search data. Models based on Google data alone, outperform the baseline models in all cases. The models achieve a reduction over the baseline models of the mean squared forecasting error for transactions and prices of up to 35 and 54 per cent, respectively.

The results suggest that Google data can serve as an early market indicator. The findings of this study suggest that the inclusion of Google search data in forecasting models can improve forecast accuracy significantly. This implies that commercial real estate forecasters should consider incorporating this free and timely data set into their market forecasts or when performing plausibility checks for future investment decisions.

This is the first paper applying Google search query data to the commercial real estate sector.

Establishing the strength of a novel variable–mortgage debt as a fraction of US gross domestic product (GDP)–on forecasting capitalisation rates in both the US office and multifamily sectors.

The authors specify a vector error correction model (VECM) to the data. VECM are used to address the nonstationarity issues of financial variables while maintaining the information embedded in the levels of the data, as opposed to their differences. The cap rate series used are from Green Street Advisors and represent transaction cap rates which avoids the problem of artificial smoothness found in appraisal-based cap rates.

Using a VECM specified with the novel variable, unemployment and past cap rates contains enough information to produce more robust forecasts than the traditional variables (return expectations and risk premiums). The method is robust both in and out of sample.

This has direct implications for governmental policy, offering a path to real estate price stability and growth through mortgage access–functions largely influenced by the Fed and the quasi-federal agencies Fannie Mae and Freddie Mac. It also offers a timely alternative to interest rate-based forecasting models, which are likely to be less useful as interest rates are to be held low for the foreseeable future.

This study offers a new and highly explanatory variable to the literature while being among the only to model either (1) transactional cap rates (versus appraisal) (2) out-of-sample data (versus in-sample) (3) without the use of the traditional variables thought to be integral to cap rate modelling (return expectations and risk premiums).

Commercial real estate and office rental values, in particular, have long been the focus of research. Several forecasting frameworks for office rental values in multivariate and univariate fashions have been proposed. Recent developments in time series forecasting using machine learning and deep learning methods offer an opportunity to update traditional univariate forecasting frameworks.

With the aim to extend research on univariate rent forecasting a hybrid methodology combining both ARIMA and a neural network model is proposed to exploit the unique strengths of both methods in linear and nonlinear modelling. N-BEATS, a deep learning algorithm that has demonstrated state-of-the-art forecasting performance in major forecasting competitions, are explained. With the ARIMA model, it is jointly applied to the office rental dataset to produce forecasts for four-quarters ahead.

When the approach is applied to a dataset of 21 major European office cities, the results show that the ensemble model can be an effective approach to improve the prediction accuracy achieved by each of the models used separately.

Real estate forecasting is essential for assessing the value of managing portfolios and for evaluating investment strategies. The approach applied in this paper confirms the heterogeneity of real estate markets. The application of mixed modelling via linear and nonlinear methods decreases the uncertainty of abrupt changes in rents.

To the best of the authors' knowledge, no such application of a hybrid model updating classical statistical forecasting with a deep learning neural network approach in the field of commercial real estate rent forecasting has been undertaken.

This study aims to provide a comprehensive analysis of various approaches to the residential property market evolution modelling and to examine the macroeconomic fundamentals that have shaped this market development in Ukraine in recent years.

The study uses a comprehensive data set encompassing relevant macroeconomic indicators and historical apartment prices. Multifactor linear regression (MLR) and ridge regression (RR) models are constructed to identify the impact of multiple predictors on apartment prices. Additionally, the ARIMAX model integrates time series analysis and external factors to enhance modelling and forecasting accuracy.

The investigation reveals that MLR and RR yield accurate predictions by considering a range of influential variables. The hybrid ARIMAX model further enhances predictive performance by fusing external indicators with time series analysis. These findings underscore the effectiveness of a multidimensional approach in capturing the complexity of housing price dynamics.

This research contributes to the real estate modelling and forecasting literature by providing an analysis of multiple linear regression, RR and ARIMAX models within the specific context of property price prediction in the turbulent Ukrainian real estate market. This comprehensive analysis not only offers insights into the performance of these methodologies but also explores their adaptability and robustness in a market characterized by evolving dynamics, including the significant influence of external geopolitical factors.

Recent research has found significant relationships between internet search volume and real estate markets. This paper aims to examine whether Google search volume data can serve as a leading sentiment indicator and are able to predict turning points in the US housing market. One of the main objectives is to find a model based on internet search interest that generates reliable real-time forecasts.

Starting from seven individual real-estate-related Google search volume indices, a multivariate probit model is derived by following a selection procedure. The best model is then tested for its in- and out-of-sample forecasting ability.

The results show that the model predicts the direction of monthly price changes correctly, with over 89 per cent in-sample and just above 88 per cent in one to four-month out-of-sample forecasts. The out-of-sample tests demonstrate that although the Google model is not always accurate in terms of timing, the signals are always correct when it comes to foreseeing an upcoming turning point. Thus, as signals are generated up to six months early, it functions as a satisfactory and timely indicator of future house price changes.

The results suggest that Google data can serve as an early market indicator and that the application of this data set in binary forecasting models can produce useful predictions of changes in upward and downward movements of US house prices, as measured by the Case–Shiller 20-City House Price Index. This implies that real estate forecasters, economists and policymakers should consider incorporating this free and very current data set into their market forecasts or when performing plausibility checks for future investment decisions.

This is the first paper to apply Google search query data as a sentiment indicator in binary forecasting models to predict turning points in the housing market.

This research developed a model to understand and predict housing market dynamics and evaluate the significance of house permits data in the model’s forecasting capability.

The research uses a multilevel algorithm consisting of a machine-learning regression model to predict the independent variables and another regressor to predict the dependent variable using the forecasted independent variables.

The research establishes a statistically significant relationship between housing permits and house prices. The novel approach discussed in this paper has significantly higher prediction capabilities than a traditional regression model in forecasting monthly average prices (R-squared value: 0.5993), house price index prices (R-squared value: 0.99) and house sales prices (R-squared value: 0.7839).

The impact of supply, demand and socioeconomic factors will differ in various regions. The forecasting capability and significance of the independent variables can vary, but the methodology can still be applicable when provided with the considered variables in the model.

The resulting model is helpful in the decision-making process for investments, house purchases and construction as the housing demand increases across various cities. The methodology can benefit multiple players, including the government, real estate investors, homebuyers and construction companies.

Existing algorithms and models do not consider the number of new house constructions, monthly sales and inventory in the real estate market, especially in the United States. This research aims to address these shortcomings using current socioeconomic indicators, permits, monthly real estate data and population information to predict house prices and inventory.

The purpose of this paper is twofold: the authors initially survey a sample of literature published after the Great Recession that address macroeconomic and commercial real estate forecasting methods related to the Great Recession and compare significant lessons learned, or lack thereof. The authors then seek to identify new models to improve the predictability of commercial real estate early warning signals regarding cyclical turning points which result in negative appreciation rates.

The authors develop a probit model to estimate quarterly probabilities of negative office appreciation returns using an alternative methodology to Tsolaco et al. (2014). The authors’ alternative method incorporates generally publicly available macroeconomic and real estate variables such as gross domestic product, office-related employment sectors, cap rate spreads, and commercial mortgage flow of funds into a probit model in order to estimate the probability of future quarterly negative office appreciation rates.

The authors’ models demonstrate the predictive power of macroeconomic variables typically associated with office demand. The probit model specification shows probabilities of negative office appreciations rates greater than 50 percent either as the quarterly office returns become negative, or in some cases several quarters before office returns become negative, for both the Great Recession and the recession occurring in the early 1990s. The models fail to show probabilities greater than 50 percent of negative office returns until after they occur for the recession in 2001. While this indicates need for further improvement in early warning models, the models do predict the more severe periods of negative office returns in advance, indicating the findings useful to real estate investors to monitor the changes in economic and real estate data identified as statistically significant in the results.

The Great Recession is a unique laboratory of significant contractions, recessions, and recoveries that challenge pre-recessionary real estate cycle models. The models provide guidance on which historical economic indicators are important to track, and gives a framework with which to calculate the probability that office prices are likely to decline. Because the models use macroeconomic indicators that are publicly available from at least one quarter in the past, the models or variations of them may provide real estate professionals with some indication of an impending decrease in office prices, even if that indication comes only one quarter in advance. Armed with this information, property owners, investors, and brokers can make more informed decisions on whether to buy or sell, and how sensitive their real estate transactions may be to timing.

The authors introduce several new models that examine the ability of historical macroeconomic indicators to provide early warning signals and identify turning points in real estate valuations, specifically negative office appreciation rates caused by the Great Recession. Using data from at least one quarter in the past, all the data in the models are publicly available (excluding National Council of Real Estate Investment Fiduciaries data) at the observed return quarter being predicted, which gives practitioners rational insights that can provide at least one source of guidance about the likelihood of an impending decrease in office prices.

The purpose of this paper is to test the asymptotic properties and prediction accuracy of two innovative methods proposed along the hedonic debate: the geographically weighted regression (GWR) and the generalized additive model (GAM).

The authors assess the asymptotic properties of linear, spatial and non-linear hedonic models based on a very large data set in Germany. The employed functional form is based on the OLS, GWR and the GAM, while the estimation methodology was chosen to be iterative in forecasting, the fitted rents for each quarter based on their 1-quarter-prior functional form. The performance accuracy is measured by traditional indicators such as the error variance and the mean squared (percentage) error.

The results provide evidence for a clear disadvantage of the GWR model in out-of-sample forecasts. There exists a strong out-of-sample discrepancy between the GWR and the GAM models, whereas the simplicity of the OLS approach is not substantially outperformed by the GAM approach.

For policymakers, a more accurate knowledge on market dynamics via hedonic models leads to a more precise market control and to a better understanding of the local factors affecting current and future rents. For institutional researchers, instead, the findings are essential and might be used as a guide when valuing residential portfolios and forecasting cashflows. Even though this study analyses residential real estate, the results should be of interest to all forms of real estate investments.

Sample size is essential when deriving the asymptotic properties of hedonic models. Whit this study covering more than 570,000 observations, this study constitutes – to the authors’ knowledge – one of the largest data sets used for spatial real estate analysis.

The Basel Committee regulations require the estimation of value-at-risk (VaR) at 99 percent confidence level for a ten-trading-day-ahead forecasting horizon. The paper provides a multivariate modelling framework for multi-period VaR estimates for leptokurtic and asymmetrically distributed real estate portfolio returns. The purpose of the paper is to estimate accurate ten-day-ahead 99%VaR forecasts for real estate markets along with stock markets for seven countries across the world (the USA, the UK, Germany, Japan, Australia, Hong Kong and Singapore) following the Basel Committee requirements for financial regulation.

A 14-dimensional multivariate Diag-VECH model for seven equity indices and their relative real estate indices is estimated. The authors evaluate the VaR forecasts over a period of two weeks in calendar time, or ten-trading-days, and at 99 percent confidence level based on the Basle Committee on Banking Supervision requirements.

The Basel regulations require ten-day-ahead 99%VaR forecasts. This is the first study that provides successful evidence for ten-day-ahead 99%VaR estimations for real estate markets. Additionally, the authors provide evidence that there is a statistically significant relationship between the magnitude of the ten-day-ahead 99%VaR and the level of dynamic correlation for real estate and stock market indices; a valuable recommendation for risk managers who forecast risk across markets.

Risk managers, investors and financial institutions require dynamic multi-period VaR forecasts that will take into account properties of financial time series. Such accurate dynamic forecasts lead to successful decisions for controlling market risks.

This paper is the first approach which models simultaneously the volatility and VaR estimates for real estate and stock markets from the USA, Europe and Asia-Pacific over a period of more than 20 years. Additionally, the local correlation between stock and real estate indices has statistically significant explanatory power in estimating the ten-day-ahead 99%VaR.