University of Twente Student Theses
The impact of different types of urban green environments on property value (Alkmaar, Netherlands)
Fallahianbizhan, Majedeh (2024) The impact of different types of urban green environments on property value (Alkmaar, Netherlands).
|
PDF
6MB |
| Abstract: | The rate of urbanization and population growth is increasing rapidly worldwide. It is estimated that a significant portion of the world's population will live in urban areas by 2050. According to this phenomenon, the demand for properties has increased in cities to accommodate the growing population that pressures the property market. Simultaneously, various factors should be considered when estimating the value of properties. This research aims to investigate the effects of different types of green environments on property value. The factors affecting the property value are categorized into three groups: locational, physical, and environmental characteristics of the property. Among different categories affecting property value, one of the most critical factors is the distance to green environments. However, there are different types of green environments with specific characteristics across the cities that are not in similar conditions. Thus, the main research problem lies in understanding how and to what extent different types of green environments, along with other factors, affect property value. This research provides empirical evidence on how different types of urban green environments influence property value, helping urban planners and policymakers make informed urban development decisions. Regarding the methodology, this research elaborated on the quantitative property valuation method to find the effects of different types of green environments on property values. First, urban green environments are classified based on size, height, density, type of vegetation, and services they provide. Then, property value prediction models are constructed by combining two-dimensional (2D) factors, for instance, the size of the property and distance to CBD, with three-dimensional (3D) factors, such as property visibility and orientation. For 2D data, three methods, Random Forest (RF), Ordinary Least Square (OLS), and Geographically Weighted Regression (GWR), are applied. In addition, for the 3D data, the OLS method is executed for modelling. Comparing the results of applying the OLS, GWR, and RF methods illustrates that the RF explains 83.1% of the property value variation based on the adjusted R-square value, which is higher than the OLS and GWR. Hence, RF is the most suitable method to predict the property value of Alkmaar. The RF illustrates that there is a non-linear correlation between the property value and different types of green environments. For instance, the size of the green environment factor is the most important factor in the model. When the size of green environments in a distance of 25m around the properties increases from 600m2 to 800m2, the property value decreases significantly. The OLS model of property value by the 3D factors has the adjusted R square value of 0.169, meaning that the model explains the 16.9% of the property value variation. The vegetation in front of the building and the view of vegetation variables are two important factors that positively correlate with the property value. The importance of different types of urban green environments is different in each of the RF and OLS models. The RF and OLS models are also validated by k-fold cross-validation using the actual value and predicted propety value by models. The percentage error of the RF model was 12.1%, while the percentage error of the OLS model was 17.01%, indicating that the developed model with 2D data is more accurate than the model with only 3D data. |
| Item Type: | Essay (Master) |
| Faculty: | ITC: Faculty of Geo-information Science and Earth Observation |
| Subject: | 74 (human) geography, cartography, town and country planning, demography |
| Programme: | Geoinformation Science and Earth Observation MSc (75014) |
| Link to this item: | https://purl.utwente.nl/essays/102082 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page