How much position still influences the market value of a property after the last years’ events according to an Optimized Multivariate Polynomial Regression?

There are three things that matter in property: location, location, location", states Lord Harold Samuel in his iconic sentence. Even though the location is certainly not the only characteristic of a property that determines its market value, it is undoubtedly one of the most important, if not the most important. In this research, the authors wish to discuss how the fixed effects influence the best estimate of the market value of a premise through an innovative use of multi-parametric estimation techniques. In particular, the aim is to detect the different marginal appreciation of intrinsic and extrinsic characteristics of properties at three different timings: before the Covid-19 pandemic, two years after the first Covid-19 alerts but before the War in Ukraine, one year after the breakout of the War. The marginal appreciations of the building’s features are diachronically analysed for a case-study in Northern Italy through a Random Forest feature importance analysis and an Optimized Multivariate Polynomial Regression. This study also integrates several techniques, such as computer programming in Python languages, multi-parametric value assessment methods, feature selection procedures and optimisation algorithms. The results may represent an interesting ongoing monitoring of how these abnormal events affect the buyer's willingness to pay to specific characteristics of the buildings, with particular attention to the location features of the neighbourhood and accessibility.