Search Results (4)

The prediction of real estate values is vital for taxation, transactions, mortgages, and urban policy development. Values can be predicted more accurately by statistical or advanced methods together when the size of the data is huge. In metropolitan cities like İstanbul, where size of the real estate data is vast and complex, mass appraisal methods supported by Machine Learning offer a scalable and consistent alternative. This study employs six algorithms: Artificial Neural Network, Extreme Gradient Boosting, K-Nearest Neighbors, Support Vector Regression, Random Forest, and Semi-Log Regression, to estimate the values of real estate on both the Asian and European continent parts of İstanbul. In total, 168,099 residential properties were utilized along with 30 of their features from both sides of the Bosphorus. The results show that RF yielded the best performance in Beşiktaş, while XGBoost performed best in Üsküdar. ANN also produced competitive results, although slightly less accurate than those of XGBoost and RF. In contrast, traditional SVR and SLR models underperformed, especially in terms of R² and RMSE values. With its large-scale dataset, focusing on one of the greatest metropolitan areas, Istanbul, and the usage of multiple ML algorithms, this study stands as a comprehensive and practical contribution to the field of automated real estate valuation. Full article

This study concerns the analytical formulation and relative implications of bottom value (BV) and mortgage lending value (MLV) regarding properties where the existing building provides an income during its useful life, leaving thereafter only the land value. The bottom value is equal to the overall property’s market value minus all incomes not collected by the end of the building’s economic life. Furthermore, it considers the income rates for land and buildings differently according to the investment type, while the mortgage lending value considers, instead, a unique rate. The mortgage lending value assessment is conducted under restrictive assumptions on long-term aspects, future marketability, and local market conditions. For the first time, mathematical and appraisal models have been applied to determine the mortgage lending value and the bottom value in particular cases, such as that mentioned above and considered in the present study (existing buildings providing income during their useful life). Some of the indexes introduced in the paper are completely original with respect to the current reference literature. Full article

With the rapidly increasing house prices in the Netherlands, there is a growing need for more localised value predictions for mortgage collaterals within the financial sector. Many existing studies focus on modelling house prices for an individual city; however, these models are often not interesting for mortgage lenders with assets spread out all over the country. That is why, with the current abundance of national geospatial datasets, this paper implements and compares three hedonic pricing models (linear regression, geographically weighted regression, and extreme gradient boosting—XGBoost) to model real estate appraisals values for five large municipalities in different parts of the Netherlands. The appraisal values used to train the model are provided by Stater N.V., which is the largest mortgage service provider in the Netherlands. Out of the three implemented models, the XGBoost model has the highest accuracy. XGBoost can explain 83% of the variance with an RMSE of €65,312, an MAE of €43,625, and an MAPE of 6.35% across the five municipalities. The two most important variables in the model are the total living area and taxation value, which were taken from publicly available datasets. Furthermore, a comparison is made between indexation and XGBoost, which shows that the XGBoost model is able to more accurately predict the appraisal values of different types of houses. The remaining unexplained variance is most probably caused by the lack of good indicators for the condition of the house. Overall, this paper highlights the benefits of open geospatial datasets to build a national real estate appraisal model. Full article

As house mortgage appraisal values have played a leading role in the 2007–2012 financial crisis, it is important to develop robust mass appraisal models that correctly estimate these values. The present paper intends to propose a methodology to examine the spatial distribution of house mortgage appraisal values. To do so, we analyzed the effect that these values, cadastral urban land values, characteristics of houses, and socioeconomic conditions and services in neighborhoods, have on house mortgage appraisal values in the 70 boroughs of Valencia (Spain). Econometric and spatial models were used, and variables were calculated as the mean and weighted values per boroughs. Our results showed that the hierarchy of cadastral values impacted mortgage appraisal values. Conversely, not all the boroughs-related variables influenced the mean mortgage values of houses, although some did anomalously. We conclude that the spatial error or autoregressive models provided very good fit results, which somewhat improved the ordinary least square model. Moreover, house mortgage appraisal values may be influenced by not only cadastral values but also by some district characteristics like mean family property size, vehicle age, distance from a metro station or from infant or primary education centers. Full article