The esteemed conference was organized by the Athens University of Economics and Business (AUEB). The topics ranged from real estate econometrics and valuation methods to urban and regional development and machine learning approaches in real estate analysis. The participants expressed their special thanks to the conference chairs, Prof. Dr. Andrianos Tsekrekos and Prof. Dr. Manolis Kavussanos, for the excellent organization and the scientifically and culturally successful event.

In the pre-event to the main conference, young researchers from all over the world came together for a PhD seminar. In this context, Jan Schmid presented his research project entitled ‘A Random Forest Meta-Learning Approach for Optimal AI Algorithm Selection in Real Estate Market Prediction’. The discussion with colleagues from various disciplines enabled an intensive methodological exchange and opened up new perspectives for the further development of the project.

In the main programme, Jan Schmid also presented the paper ‘Predicting House Price Indices: A Machine Learning Approach Using Linked Listing and Transaction Data’. The study focuses on predicting real estate price indices based on micro- and macroeconomic data using combined machine learning methods. The methodology presented met with broad interest among the expert audience and was the starting point for numerous responses that provided new impetus for ongoing research.

Also, in the main programme, Dr. De Souza presented his research entitled 'The Baulandumlegung Prognosis Algorithm: A Non-Stationary Spatiotemporal Approach for Forecasting Land Readjustment Projects Using Machine Learning and 120 Years of Unstructured Data from Frankfurt am Main (1902–2022)'. The study introduced a novel forecasting methodology combining hedonic spatial regression, Mahalanobis distance matching, difference-in-differences estimation, anisotropic variogram analysis, and a historical weight function informed by historical institutionalism. Dr. De Souza's work uniquely integrates historical archival data spanning over a century with advanced spatial statistical methods to organize and classify land readjustment project outcomes. His presentation generated discussions about the applicability of machine learning techniques and historical data analysis in real estate economics.

In addition to scientific deepening, the conference also provided space for international networking. New contacts were made, opportunities for cooperation explored, and existing research approaches reflected upon in an open exchange. ERES 2025 once again underlined the growing importance of interdisciplinary and data-based approaches in real estate economics – and the role that European universities such as TU Darmstadt are playing in this transformation.