1. Introduction
Multi-parametric assessment methods are particularly interesting in property valuation [
1] since they are able to define all the factors that contribute to the determination of market prices, while also identifying the marginal influence that each factor has on the price [
2]. In this context, we distinguish construction and positional factors. Among the
construction characteristics, there are, for instance, features like the dimension of the building, the floor level, the state of maintenance, the presence of installations, domotics, balconies, gardens, or parking spaces. On the other hand,
location characteristics refer to the position in a central/semi-central/suburban area of a city, the proximity to commercial facilities and services, the access to public transport, as well as the neighbourhood quality. Such positional factors are also defined as fixed effects, since a property, apart from its construction characteristics, cannot be considered separately from the positional context in which it is placed [
3]. Such fixed effects may have both a positive [
4] or a negative [
5] influence on prices depending on the market segment considered.
In this study, the scope is to examine the impact that fixed effects produce on prices in comparison to construction characteristics, by means of a multi-parametric econometric analysis [
6,
7].
An econometric market value assessment application, in fact, is able to determine the relationship that links some features describing the property (the independent variables) and its market value (the dependent variable) [
8].
2. Materials and Methods
Presented in
Figure 1, the methodological strategy is aimed at producing a reliable hedonic pricing model that includes both construction and location regressors. The methodological approach is as follows:
In Phase 1, the independent variables for the econometric model are defined (construction and location characteristics), given the market value as the dependent variable.
In Phase 2 a Random Forest (RF) feature importance analysis is carried out to test the significance of the regressors on the output.
In Phase 3, the econometric model is produced, and the multiple regression is built as in Equation (1):
In the equation above, Y is the market price, Xk are the buildings characteristics, βa is the constant, βk are the coefficients, ξ is the error.
3. Results
3.1. Case Study
Padova (a city in North-Eastern Italy) is chosen to be an exemplary case study, and the consulted information is from the first half of 2023.
The observations collected to produce the forecasting model are limited to residential buildings in Padova, including new and existing apartments, detached houses, villas, lofts, or terraces located in central, semi-central, and suburban areas.
Particular attention is given to the selection of the location and construction characteristics of each property that need to be collected. The observations recorded describe 1518 buildings, with an average market value of 2150 EUR /m².
3.2. Feauture Selection
A feature importance and a consequent feature selection analysis is performed with the Random Forest (RF) regressor to calculate the importance coefficients that link the regressors to the market price. Such coefficients are summarized in
Figure 2 and
Figure 3.
3.3. Hedonic Pricing Model
A stepwise analysis is carried out as described in
Figure 4 to identify the best regression equation. Simulation “E” leads to the final equation, which is summarized in Equation (2): The R-square is 79.21%, and all the regressors satisfy the significance test with a
p-value < 0.05.
4. Discussion and Conclusions
The Random Forest analysis and the hedonic pricing model developed indicate that the most influential factors among the construction features are the area of the premises, the number of rooms, the energy performance level, and the maintenance conditions. As far as the building typologies are concerned, the most appreciated ones are apartments and multi-family villas. Among the location characteristics, the most impactful factors are access to a train station and proximity to commercial, cultural, and leisure services, as well as to medical centres.
The hedonic pricing model points out that greater proximity to the city centre produces greater market values (the higher the distance, the lower the price). The number of cultural services increases the market value of a building, producing a higher quality neighbourhood, and, also, the higher the energy performance of a building, the higher the market price.
In conclusion, this study has developed a unified procedure to understand processes affecting demand preferences and price formation mechanisms in a specific Italian real estate market, i.e., the city of Padova. In further development of this line of research, the approach could be expanded to other cities in Italy, and also applied at different times, so as to produce a diachronic comparison spread out over the territory that is able to map the willingness to pay for certain features of buildings.
Author Contributions
Conceptualization: M.S., L.G. and A.G.R.; methodology: M.S. and A.G.R.; software: M.S.; validation: L.G.; formal analysis: M.S. and A.G.R.; investigation: M.S., L.G. and A.G.R.; data curation: M.S., L.G. and A.G.R.; writing—original draft preparation: A.G.R.; writing—review and editing, M.S., L.G. and A.G.R.; visualization, A.G.R.; supervision: M.S. and L.G. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
Not applicable.
Informed Consent Statement
Not applicable.
Data Availability Statement
The raw data supporting the conclusions of this article will be made available by the authors on request.
Conflicts of Interest
The authors declare no conflicts of interest.
References
- Simonotti, M. Metodi di Stima Immobiliare; Flaccovio: Palermo, Italy, 2006. [Google Scholar]
- Gabrielli, L.; Ruggeri, A.G.; Scarpa, M. Using Artificial Neural Networks to Uncover Real Estate Market Transparency: The Market Value. In Lecture Notes in Computer Science, Proceedings of the Computational Science and Its Applications—ICCSA 2021, Cagliari, Italy 13–16 September 2021; Gervasi, O., Murgante, B., Misra, S., Garau, C., Blečić, I., Taniar, D., Apduhan, B.O., Rocha, A.M.A.C., Tarantino, E., Torre, C.M., Eds.; Springer International Publishing: Cham, Switzerland, 2021; Volume 12954, pp. 183–192. [Google Scholar]
- Adair, A.; McGreal, S.; Smyth, A.; Cooper, J.; Ryley, T. House prices and accessibility: The testing of relationships within the Belfast Urban Area. Hous. Stud. 2000, 15, 699–716. [Google Scholar] [CrossRef]
- Moreno Gil, S.; Ritchie, J.R.B. Understanding the museum image formation process: A comparison of residents and tourists. J. Travel Res. 2009, 47, 480–493. [Google Scholar] [CrossRef]
- Waddell, P.; Shukla, V. Employment Dynamics, Spatial Restructuring, and the Business Cycle. Geogr. Anal. 1993, 25, 35–52. [Google Scholar] [CrossRef]
- Wang, A.; Xu, Y. Multiple linear regression analysis of real estate price. In Proceedings of the International Conference on Robots and Intelligent System, ICRIS 2018, Changsha, China, 26–27 May 2018; IEEE: Piscataway, NJ, USA, 2018; pp. 564–568. [Google Scholar]
- Feng, J.; Zhu, J. Nonlinear regression model and option analysis of real estate price. Dalian Ligong Daxue Xuebao/J. Dalian Univ. Technol. 2017, 57, 545–550. [Google Scholar] [CrossRef]
- Štubňová, M.; Urbaníková, M.; Hudáková, J.; Papcunová, V. Estimation of residential property market price: Comparison of artificial neural networks and hedonic pricing model. Emerg. Sci. J. 2020, 4, 530–538. [Google Scholar] [CrossRef]
| Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).