Riassunto analitico
This thesis is intended to serve as an exhaustive examination of the current statistical tools and techniques employed in assessing and valuating the real estate market. Given the inherent complexity of real estate valuation, multiple strategies have been developed over the years. Each of these approaches, be it statistical, machine learning, or a combination of both, have unique strengths and limitations. The purpose of this work is to present a comprehensive analysis of these methods, understanding their benefits and challenges, and ultimately proposing an integrated approach that harnesses the best of both worlds.
In a complex environment such as Real Estate, the traditional statistical tools, while offering a degree of clarity and transparency, sometimes fall short of providing an accurate prediction of the market behavior. Conversely, the advanced machine learning approaches, although more proficient in forecasting accuracy, often suffer from a lack of interpretability, making their results less straightforward to stakeholders.
Within this context, this thesis attempts to identify the optimal balance between the predictive accuracy of machine learning models and the transparency of statistical methods. It aims to elucidate the pros and cons of each methodology, and by juxtaposing these diverse techniques, we hope to uncover how to best combine them for accurate and transparent real estate valuation and prediction.
In the quest for the most robust and reliable forecasting tool for the real estate industry, we begin this exploration with an in-depth understanding of the concepts of value, market, and valuation standards. From this foundation, we develop a comprehensive framework for real estate valuation, featuring a wide array of valuation methods.
Chapter 1 delves into the understanding of valuation for real estates through various concepts of value, market, and valuation according to the International Valuation Standards (IVS). Chapter 2 is an extensive analysis of statistical methods and machine learning techniques used in the field. Finally, the last chapter presents results from two scientific papers in the field of statistical methods and machine learning, respectively.
|
