@article {Kok202, author = {Nils Kok and Eija-Leena Koponen and Carmen Adriana Mart{\'\i}nez-Barbosa}, title = {Big Data in Real Estate? From Manual Appraisal to Automated Valuation}, volume = {43}, number = {6}, pages = {202--211}, year = {2017}, doi = {10.3905/jpm.2017.43.6.202}, publisher = {Institutional Investor Journals Umbrella}, abstract = {Real estate is the third-largest asset class for institutional investors, but determining the value of commercial real estate assets remains elusively hard. In this article, the authors provide a practical application of big data by employing a unique set of data on U.S. multifamily assets, in combination with sophisticated modeling techniques, to develop an automated, machine-based valuation model for the commercial real estate sector. The authors find strong evidence of the superiority of automated valuation models over traditional appraisals: The absolute error of the automated model is 9\%, which compares favorably against the accuracy of traditional appraisals, and the model can produce an instant value at every moment in time at a very low cost. The authors also provide evidence of the importance of using hyperlocal information on the location of an asset. The model developed in this article is directly applicable for real estate lenders and investors and has important implications for the traditional appraisal industry.TOPIC: Real estate}, issn = {0095-4918}, URL = {https://jpm.pm-research.com/content/43/6/202}, eprint = {https://jpm.pm-research.com/content/43/6/202.full.pdf}, journal = {The Journal of Portfolio Management} }