University of Twente Student Theses


Spatial determinants of Real Estate Appraisals in the Netherlands: a Machine Learning Approach

Guliker, B. Evert (2021) Spatial determinants of Real Estate Appraisals in the Netherlands: a Machine Learning Approach.

[img] PDF
Abstract:In the financial sector, there is a growing need for better localised value predictions for mortgage collaterals. Money lenders know the value of a house through an appraisal once the mortgage is approved. However, 20 years later, it is unknown how much the house is increased in value without conducting another appraisal. Still, money lenders are mandated by the Authority for the Financial Markets (AFM) to make a proper risk analysis of their portfolios. Currently, at Stater N.V., the Kadaster regional index is used to index appraisals, which give a value indication for a mortgage collateral. This generalises the price increase for all types of housing to the same regional price index. This research explores how external data sources from Kadaster, CBS and RVO can help predict appraisal values of mortgages. Three types of hedonic pricing models are build using Linear Regression (LR), Geographically Weighted Regression (GWR) and Extreme Gradient Boosting (XGBoost). XGBoost ends up being the best-performing model. Trained on appraisal values from 5 large municipalities for 2020, it is able to predict the appraisals values with an average error of 6.35%. (R^2 = 0.83, RMSE = €65,312). Ultimately, the model outperforms indexation by taking into account unique housing characteristics.
Item Type:Essay (Master)
Stater N.V., Amersfoort, Netherlands
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Business Information Technology MSc (60025)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page