Just Accepted
ORIGINAL ARTICLES
Interpretable Machine Learning for the German residential rental market – shedding light into model mechanics
Severin Bachmann
Nuremberg Research Institute for Cooperative Studies, Nuremberg, Germany.
Accepted: 2024-12-04 | Published Online: 2024-12-20
DOI: 10.36253/aestim-16351
ABSTRACT
We compare the drivers in Machine learning models and give insights into their strengths and weaknesses predicting rental prices. The study employs SHAP values to measure feature importance. The study aims to investigate linear regression, decision tree and XGBoost algorithms. The research is unique in its application of IML methods to a large dataset of over 2.4 million observations in the German rental market and its application of comparative statistics using aggregate SHAP values. Main results are the superiority of XGB and LR showing higher SHAP values overall and thus explaining its lower predictive efficacy. DT models capture intricate interactions among variables with fewer features, while XGB accommodates more variables, emphasizing its higher complexity and thus superior performance. The top ten features for DT and XGB models show significant overlap, indicating robust concordance. Specific features are identified that distinguish the models, suggesting that a more complex model, like XGB, handles dummy variables more adeptly.
Alternative methods for measuring the influence of location in hedonic pricing models
Marco Aurelio Stumpf Gonzalez1,*, Diego Alfonso Erba2
1 Polytechnic School, Universidade do Vale do Rio dos Sinos, São Leopoldo, Brazil
2 Independent researcher and consultant
Accepted: 2024-11-26 | Published Online: 2024-12-02
DOI: 10.36253/aestim-15472
ABSTRACT
The effects of location play a crucial role in the real estate market, encompassing aspects of accessibility and neighborhood. However, these are elements that are not directly measurable. There are traditional ways to consider location, usually through subjective measures based on professional experience, through proxy variables. Understanding these elements is vital for estimating real estate values, whether for legal, commercial, or tax purposes. Furthermore, seeking more objective options is a relevant issue to broaden the justification of estimated values and to enable the development of mass appraisal models. This article proposes and evaluates alternative solutions based on statistics, machine learning, and geostatistics to estimate location. A study was conducted using market data from Novo Hamburgo, southern Brazil, verifying the feasibility of the options presented. Satisfactory statistical results demonstrate the viability of the proposed approach.
Integrating the Capital Asset Pricing Model with the Analytic Hierarchy Process and the Delphi Method: a proposed method for estimating the discount rate in constrained real estate development initiatives
Fabrizio Battisti1,*, Giovanna Acampa1, Mariolina Grasso2
1 Department of Architecture DIDA, University of Florence, Italy
2 Department of Engineering and Architecture, University of Enna “Kore”, Italy
Accepted: 2024-10-10 | Published Online: 2024-11-12
DOI: 10.36253/aestim-16262
ABSTRACT
The legislative framework on territorial and urban planning has become increasingly rich and complex in the European Union, particularly Italy. The structured – and often hindering – system of division of responsibilities between the central State, Regions, local institutions, and organisms generates different levels of administrative verification. The environmental and landscape constraints by which each Public Administration with jurisdiction over the territory exercises its powers strongly impact territorial management and negatively affect investments. Over the years, this has been one of the main reasons behind the significant dilation of the risk and the time required to obtain the necessary authorizations to start construction, producing “business risk.” Based on this premise, this work presents a methodological investigation of the relationships between environmental and landscape constraints, the regulatory framework involving the building, and its Market Value. This is finally aimed at finding suitable methods and procedures to formulate a reasonable discount rate considering the constraints and the related regulations that operate on an asset. A multi-step method integrating the Capital Asset Pricing Model, Analytic Hierarchy Process, and Delphi Method is proposed to assess the discount rate component related to urban risk.
Filling the old with new life. Application of original indicators for evaluating ecovillages as village repopulation initiatives
Giovanna Acampa1,*, Alessio Pino2
1 Department of Architecture DIDA, University of Florence, Italy
2 Department of Engineering and Architecture, University of Enna “Kore”, Italy
Accepted: 2024-10-10 | Published Online: 2024-10-24
DOI: 10.36253/aestim-16404
ABSTRACT
The recently intensified trend of centripetal movements from small to bigger centres has multiplied the number of inhabitants of large cities. In Italy, this has resulted in worrying figures: more than 70% of Italian Municipalities have less than 5,000 inhabitants. Despite several regional and national policies dedicating programs and funds to counteract this progressive phenomenon fostering the repopulation of abandoned villages, this trend is far from being halted. Though the functional gap between cities and villages is evident, this study and previous research on this theme aim to change the perspective on the possible uses and repopulation processes of villages, pivoting on their potential as places where to enjoy different lifestyles. The focus is on the ecovillage model, developing a set of specific indicators to individuate them through their peculiar aspects and assess their benefits and vulnerabilities. An experimental application is also proposed on 7 ecovillages. This set of indicators is not conceived as completely substitutive of those used in current policies, but rather as a suggestion of possible integrations to avoid demoting this category of villages in policy-related evaluations for funding allocation.
Urban green infrastructure valuation: an economic method for the aesthetic appraisal of hedges
Andrea Dominici1,*, Sandro Sacchelli2
1 Department of Land, Environment, Agriculture and Forestry (TESAF), University of Padova, Italy
2 Department of Agriculture, Food, Environment and Forestry (DAGRI), University of Florence, Italy
Accepted: 2024-09-12 | Published Online: 2024-09-12
DOI: 10.36253/aestim-16603
ABSTRACT
The paper presents a parametric approach to quantify the economic value of hedges in urban green spaces. The model integrates indexes that allow for an aesthetic estimate of green infrastructure. Both field and desk phases are developed to depict and sample hedgerows in a case study in Italy (Cascine Park, Florence). Street view and Google Maps applications are used in the preliminary steps to spatialize hedges. An equation, incorporating nine variables including financial, dendrometric, and correction factors, is developed to appraise economic value. The results highlight the relevance of species, plant height, and the number of hedge rows for the unitary and total value of green infrastructures. Phytosanitary condition, the presence of gaps in linear traits, and the degree of tree canopy coverage also influence the economic performances of hedges. The technique facilitates application for both researchers and practitioners, potentially allowing for damage estimates and calibrated management of urban green in different locations.
Comparing traditional and machine learning techniques in apartments mass appraisal in Fortaleza, Brazil
Antônio Augusto Ferreira de Oliveira1, Fabián Reyes-Bueno2, Marco Aurelio Stumpf González3,*, Everton da Silva4
1 Municipal treasury auditor of Fortaleza, Brazil
2 Facultad de Ciencias Exactas y Naturales, Universidad Técnica Particular de Loja, Loja, Ecuador
3 Polytechnic School, Universidade do Vale do Rio dos Sinos, São Leopoldo, Brazil
4 Geosciences Department, Universidade Federal de Santa Catarina, Florianópolis, Brazil
Accepted: 2024-07-02 | Published Online: 2024-08-01
DOI: 10.36253/aestim-15344
ABSTRACT
Mass appraisal has significant applications, such as urban planning, real estate appraisal, and property tax. Due to the challenges of analyzing massive models, they are often developed using semi-automatic assessment methods and machine learning techniques. This article explores different appraisal model methods that utilize statistics and machine learning. It also looks at incorporating spatial information to see if the chosen method can effectively capture the typical spatial dependency of the real estate market. This can help reduce the spatial autocorrelation observed in the residuals. The study compared nine machine learning methods with traditional statistical approaches using a dataset of over 43,000 apartments in Fortaleza, Brazil. The results of the machine learning algorithms were similar. The XGBoost minimized spatial autocorrelation. The easiest interpretations were with MRA, M5P, and MARS techniques. Although, these techniques had the greatest residual spatial autocorrelations. There is a trade-off between the methods, depending on whether the aim is to improve accuracy or provide a clear explanation for property taxation.