Search results

The purpose of this paper is to present an algorithm of real estate mass appraisal in which the impact of attributes (real estate features) is estimated by inequality restricted least squares (IRLS) model.

This paper presents the algorithm of real estate mass appraisal, which was also presented in the form of an econometric model. Vital problem related to econometric models of mass appraisal is multicollinearity. In this paper, a priori knowledge about parameters is used by imposing restrictions in the form of inequalities. IRLS model is therefore used to limit negative consequences of multicollinearity. In ordinary least squares (OLS) models, estimator variances might be inflated by multicollinearity, which could lead to wrong signs of estimates. In IRLS models, estimators efficiency is higher (estimator variances are lower), which could result in better appraisals.

The final effect of the analysis is a vector of the impact of real estate attributes on their value in the mass appraisal algorithm. After making expert corrections, the algorithm was used to evaluate 318 properties from the test set. Valuation errors were also discussed.

Restrictions in the form of inequalities were imposed on the parameters of the econometric model, ensuring the non-negativity and monotonicity of real estate attribute impact. In case of real estate, variables are usually correlated. OLS estimators are then inflated and inefficient. Imposing restrictions in form of inequalities could improve results because IRLS estimators are more efficient. In the case of results inconsistent with theoretical assumptions, the real estate mass appraisal algorithm enables having the obtained results adjusted by an expert. This can be important for low quality databases, which is often the case in underdeveloped real estate markets. Another reason for expert correction may be the low efficiency of a given real estate market.

Our research aims in designation of a hybrid approach in the calibration of an attribute impact vector in order to guarantee its completeness in case when other approaches cannot ensure this.

Real estate mass appraisal aims at valuating a large number of properties by means of a specialised algorithm. We can apply various methods for this purpose. We present the Szczecin Algorithm of Real Estate Mass Appraisal (SAREMA) and the four methods of calibration of an attribute impact vector. Eventually, we present its application on the example of 318 residential properties in Szczecin, Poland.

We compare the results of appraisals obtained with the application of the hybrid approach with the appraisals obtained for the three remaining ones. If the database is complete and reliable, the econometric and statistical approaches could be recommended because they are based on quantitative measures of relationships between the values of attributes and properties' unit values. However, when the database is incomplete, the expert and, subsequently, hybrid approaches are used as supplementary ones.

The application of the hybrid approach ensures that the calibration system of an attribute impact vector is always complete. This is because it incorporates the expert approach that can be used even if the database excludes application of approaches that are based on quantitative measures of relationship between the unit real estate value and the value of attributes.

This paper aims to test and compare two innovative methodologies (utility additive and evolutionary polynomial regression) for mass appraisal of residential properties. The aim is to deepen their characteristics, by exploring the potentialities and the operating limits.

With reference to the same case studies, concerning samples of residential properties recently sold in three Italian cities, the two procedures are tested and the results are compared. The first method is the utility additive, which interprets the process of the property price formation as a multi-criteria selection of multi-objective typology, where the selection criteria are the property characteristics that are decisive in the real estate market; the second method is a hybrid data-driven technique, called evolutionary polynomial regression, that uses multi-objective genetic algorithms to search those models expressions that simultaneously maximize accuracy of data and parsimony of mathematical functions.

The outputs obtained from the experimentation highlight the potentialities and the limits of the two methodologies, as well as the possibility of jointly applying them to interpret and predict the real estate phenomena in a more realistic representation.

In all countries, mass appraisal techniques have become strategic for the definition of management and enhancement policies of public and private property assets, in the case of investments of technical and economic refunctionalization (energy, environment, etc.), and for the alienation of buildings no longer suitable for public needs (military barracks, hospitals, areas in disuse, etc.). In this context, the use of mass appraisal techniques for residential properties assumes a leading role for sector operators (buyers, sellers, institutions, insurance companies, banks, real estate funds, etc.). Therefore, the results of the applications outline the potentialities of the two methodologies implemented and the opportunity of further insights of the topics that have been dealt with in this research.

Real estate appraisals are becoming an increasingly important means of backing up financial operations based on the values of these kinds of assets. However, in very large databases, there is a reduction in the predictive capacity when traditional methods, such as multiple linear regression (MLR), are used. This paper aims to determine whether in these cases the application of data mining algorithms can achieve superior statistical results. First, real estate appraisal databases from five towns and cities in the State of Paraná, Brazil, were obtained from Caixa Econômica Federal bank.

After initial validations, additional databases were generated with both real, transformed and nominal values, in clean and raw data. Each was assisted by the application of a wide range of data mining algorithms (multilayer perceptron, support vector regression, K-star, M5Rules and random forest), either isolated or combined (regression by discretization – logistic, bagging and stacking), with the use of 10-fold cross-validation in Weka software.

The results showed more varied incremental statistical results with the use of algorithms than those obtained by MLR, especially when combined algorithms were used. The largest increments were obtained in databases with a large amount of data and in those where minor initial data cleaning was carried out. The paper also conducts a further analysis, including an algorithmic ranking based on the number of significant results obtained.

The authors did not find similar studies or research studies conducted in Brazil.

The purpose of this paper is to apply boosted regression trees (BRT) to a heterogeneous data set of residential property drawn from a jurisdiction in Malaysia, with the objective to evaluate its application within the mass appraisal environment in Malaysia. Machine learning (ML) techniques have been applied to real estate mass appraisal with varying degrees of success.

To evaluate the performance of the BRT model two multiple regression analysis (MRA) models have been specified (linear and non-linear). One of the weaknesses of traditional regression is the need to a priori specify the functional form of the model and to ensure that all non-linearities have been accounted for. For a BRT model the algorithm does not require any predetermined model or variable transformations, making the process much simpler.

The results show that the BRT model outperformed the MRA-specified models in terms of the coefficient of dispersion and mean absolute percentage error. While the results are encouraging, BRT models still lack transparency and suffer from the inability to translate variable importance into quantifiable variable effects.

This paper presents a useful alternative modelling technique, BRT, for use within the mass appraisal environment in Malaysia. Its advantages include less intensive data cleansing, no requirement to specify the predictive underlying model, ability to utilise categorical variables without the need to transform them and not as data hungry, as for example, MRA.

This paper adds to the knowledge in this area by applying a relatively new ML model, BRT to residential property data from a jurisdiction in Malaysia. BRT has shown promise as a strong predictive model when applied in other disciplines; therefore this research empirically tests this finding within real estate valuation.

As regards the assessment of the market values of properties that compose real estate portfolios, the purpose of this paper is to propose and test an automated valuation model. In particular, the method defined allows for providing for objective, reliable and “quick” valuations of the assets in the phases of periodic reviews of the property values.

Aiming at both predictive and interpretative purposes, the method, based on multi-objective genetic algorithms to search those model expressions that simultaneously maximize the accuracy of the data and the parsimony of the mathematical functions, is applied to a sample data of office properties characterized by medium and large size, located in the city of Milan (Italy) and sold in the period between 2004 and 2015.

The model obtained could be an integration of the canonical methodologies (market approach, income approach, cost approach) implemented in the assessment of the market values of properties, so as to provide an additional tool to verify the results. In particular, the inclusion of economic variables in the model is consistent with the need to reiterate the valuations, contextualizing them to the locational characteristics and to the current property cycle phase in the specific area.

The model can be applied by all the operators involved in the periodic reviews of the values of property portfolios: from real estate funds’ insiders, in order to monitor the values obtained through the canonical approaches, to the public institutions, such as the revenue agencies, in order to ensure the fair payment of the taxes through the updating values of the properties according to the actual and current market trends.

The method proposed can be a valid support for all public and private entities that hold significant property assets and that, for various reasons (periodic reviews of the balance sheets, sales, enhancement, investment, etc.), require cyclical updated values of the properties. The automated valuation model developed can be used for the assessment of “comparison” values with the estimates values obtained by other assessment techniques, in order to ensure a further monitoring tool of the results from the subjects involved.

The paper aims to investigate the application of particle swarm optimisation and back propagation in weights optimisation and training of artificial neural networks within the mass appraisal industry and to compare the performance with standalone back propagation, genetic algorithm with back propagation and regression models.

The study utilised linear regression modelling before the semi-log and log-log models with a sample of 3,242 single-family dwellings. This was followed by the hybrid systems in the selection of optimal attribute weights and training of the artificial neural networks. Also, the standalone back propagation algorithm was used for the network training, and finally, the performance of each model was evaluated using accuracy test statistics.

The study found that combining particle swarm optimisation with back propagation in global and local search for attribute weights enhances the predictive accuracy of artificial neural networks. This also enhances transparency of the process, because it shows relative importance of attributes.

A robust assessment of the models’ predictive accuracy was inhibited by fewer accuracy test statistics found in the software. The research demonstrates the efficacy of combining two models in the assessment of property values.

This work demonstrated the practicability of combining particle swarm optimisation with back propagation algorithms in finding optimal weights and training of the artificial neural networks within the mass appraisal environment.

The purpose of this paper is to present how prior knowledge about the impact of real estate features on value might be utilised in the econometric models of real estate appraisal. In these models, price is a dependent variable and real estate features are explanatory variables. Moreover, these kinds of models might support individual and mass appraisals.

A mixed estimation procedure was discussed in the research. It enables using sample and prior information in an estimation process. Prior information was provided by real estate experts in the form of parameter intervals. Also, sample information about the prices and features of undeveloped land for low-residential purposes was used. Then, mixed estimation results were compared with ordinary least squares (OLS) outcomes. Finally, the estimated econometric models were assessed with regard to both formal criteria and valuation accuracy.

The OLS results were unacceptable, mostly because of the low quality of the database, which is often the case on local, undeveloped real estate markets. The mixed results are much more consistent with formal expectations and the real estate valuations are also better for a mixed model. In a mixed model, the impact of each real estate feature could be estimated, even if there is no variability in the sample information. Valuations are also more precise in terms of their consistency with market prices. The mean error (ME) and mean absolute percentage error (MAPE) are lower for a mixed model.

The crucial problem in econometric property valuation is that it involves the unreliability of databases, especially on undeveloped, local markets. The applied mixed estimation procedure might support sample information with prior knowledge, in the form of stochastic restrictions imposed on parameters. Thus, that kind of knowledge might be obtained from real estate experts, practitioners, etc.

The traditional models of real estate market have several sources of imprecision, such as transitions between submarkets, generating difficulties in property valuation. The purpose of this paper is to examine an alternative to improve mass appraisal models, using fuzzy rules.

Fuzzy rule‐based systems (FRBS) are able to generate flexible systems and may be useful in considering vagueness or imprecision presents in real estate market. An application to the housing market of Porto Alegre (Brazil), with more than 30,000 apartments, transacted in 1998‐2001, illustrates the fuzzy system, comparing with traditional hedonic regression model.

The results have indicated the potential of fuzzy rules to use in mass appraisal.

This paper presents a procedure to develop mass appraisal models using FRBS.

In the literature there are numerous tests that compare the accuracy of automated valuation models (AVMs). These models first train themselves with price data and property characteristics, then they are tested by measuring their ability to predict prices. Most of them compare the effectiveness of traditional econometric models against the use of machine learning algorithms. Although the latter seem to offer better performance, there is not yet a complete survey of the literature to confirm the hypothesis.

All tests comparing regression analysis and AVMs machine learning on the same data set have been identified. The scores obtained in terms of accuracy were then compared with each other.

Machine learning models are more accurate than traditional regression analysis in their ability to predict value. Nevertheless, many authors point out as their limit their black box nature and their poor inferential abilities.

AVMs machine learning offers a huge advantage for all real estate operators who know and can use them. Their use in public policy or litigation can be critical.

According to the author, this is the first systematic review that collects all the articles produced on the subject done comparing the results obtained.