House Price Modeling over Heterogeneous Regions with Hierarchical Spatial Functional Analysis

Abstract

Online real-estate information systems such as Zillow and Trulia have gained increasing popularity in recent years. One important feature offered by these systems is the online home price estimate through automated data-intensive computation based on housing information and comparative market value analysis. State-of-the-art approaches model house prices as a combination of a latent land desirability surface and a regression from house features. However, by using uniformly damping kernels, they are unable to handle irregularly shaped regions or capture land value discontinuities within the same region due to the existence of implicit sub-communities, which are common in real-world scenarios. In this paper, we explore the novel application of recent advances in spatial functional analysis to house price modeling and propose the Hierarchical Spatial Functional Model (HSFM), which decomposes house values into land desirability at both the global scale and hidden local scales as well as the feature regression component. We propose statistical learning algorithms based on finite-element spatial functional analysis and spatial constrained clustering to train our model. Extensive evaluations based on housing data in a major Canadian city show that our proposed approach can reduce the mean relative house price estimation error down to 6.60%.

Space Partitioning and Regression Mode Seeking via a Mean-Shift-Inspired Algorithm

“…However, the plug-in approach does not directly render an algorithm to find the local modes, which is usually challenging because the local modes are only implicitly defined through the regression estimators. As a special case of goal (2), spatial partitioning has many applications in, for example, clustering for house price (Liu et al, 2016), segregated homogeneous neighborhoods studied in sociology (Legewie, 2018), and division of disease risk zones in epidemiology (Gaudart et al 2005). The regression MS algorithm we propose in this paper uses a modal clustering idea and can be simultaneously useful for these two goals.…”

Roomsemble: Progressive web application for intuitive property search

“…One of the last domains to be disrupted by technology is real estate, with research mainly focusing on price predictions [5], [9], [10], and [12]. Although price is one of the most important factors in making a decision, it is usually not the only factor that motivates buyers.…”

Defining Geographical Rating Territories in Auto Insurance Regulation by Spatially Constrained Clustering

“…Some current research is now considering to focus on spatially-constrained clustering. In Liu et al (2016), spatially-constrained functional clustering was used to model house prices to capture differences among heterogeneous regions. In Liao and Peng (2012), an algorithm called clustering with local search (CLS) was proposed to efficiently derive clusters for certain types of spatial data.…”