Machine learning valuation I

Project title

Space-time machine learning models for valuation and prediction of real estate objects.

Project goal

The goal of this project was to investigate modern methods of data science and machine learning that can handle large amounts of data in order to develop novel non-stationary models for the pricing of real estate objects.

Time period

2018 until 2020

Research partners

Hochschule Luzern

Banque Cantonale Vaudoise

UBS

Zürcher Kantonalbank

Publications

Maximum Likelihood Estimation of Spatially Varying Coefficient Models for Large Data with an Application to Real Estate Price Prediction

Examining the Vintage Effect in Hedonic Pricing using Spatially Varying Coefficients Models: A Case Study of Single-Family Houses in the Canton of Zurich

Financing

Innosuisse – Schweizerische Agentur für Innovationsförderung

FPRE