A rapidly updating stratified mix-adjusted median property price index model

A real-time mix-adjusted median property price index enabled by an efficient nearest neighbour approximation data structure

“…As a result, when the GeoTree is performing the SCB computation, it will now only match properties which are both nearby and share the same number of bedrooms as the property in question. This allows the index model to compare the price ratio of properties which are more similar in nature during the stratification stage and thus should result in a smoother, more accurate measure of the change in property prices over time [11].…”

“…Despite this advancement, the algorithm still has great benefit to the layperson, outperforming certain implementations of hedonic regression models without having access to richer, private datasets [11]. The result is a highly flexible algorithm, which can adapt to various levels of data availability while still offering a high degree of accuracy.…”

“…In this article, we will expand upon previous work by Maguire et al [5] on a stratified, mix-adjusted median property price model by applying said algorithm to a larger and richer dataset of property listings and explore the enhancements in smoothness offered by evolving the original algorithm [11]. Such evolutions have been made possible through the introduction of a custom-tailored data structure, the GeoTree, which allows for rapid identification of a group of neighbours of any given property within fixed distance buckets [12].…”