Real Estate Price Analytics in the US Market Based on Implicit Features

This master's thesis is devoted to the analysis of the influence of implicit factors on real estate prices. The study focuses on the significant economic importance of the real estate market, which is increasing with digitalization and increased data availability. Despite this, existing valuation models often ignore implicit factors such as environmental indicators or crime rates, potentially reducing forecast accuracy and increasing risks for investors. The central hypothesis of the paper is the claim that integrating data on implicit factors into scoring models can significantly improve the quality of real estate value predictions. To test this hypothesis, various approaches to value estimation are analyzed, including both traditional methods and modern technological solutions such as automated systems and geographic information technology. In the process of proving the central hypothesis, the work was divided into 3 main semantic parts: theoretical review, analysis of implicit factors and development of a new scoring model. The first part discusses the evolution of scoring methods and their current state, pointing out their shortcomings and possible ways of improvement. The second part is devoted to the identification and analysis of implicit factors that may influence the house price index. The third part describes the development and testing of a new valuation model that incorporates the implicit factors. As a result, it is demonstrated how the proposed model can improve the accuracy of valuations and reduce investment risks. The paper proves the importance of implicit factors in real estate valuation, showing that their consideration can lead to more accurate and efficient results in investment decisions. This study represents a significant contribution to the field of real estate price analysis, emphasizing the importance of an integrated approach to achieve better forecasting results.