Using Data Science To Support Real Estate Investment Strategy
KEY HIGHLIGHTS FROM OUR OCTOBER 27, 2021 DISCUSSION ON USING DATA SCIENCE TO SUPPORT REAL ESTATE INVESTMENT STRATEGY
GRI was pleased to have hosted Erkan Yonder from John Molson School of Business, Concordia University, on October 27th 2021, who shared highlights from his recent work titled, “Neighborhood Effects, Immigration and Real Estate Valuation: A Machine Learning Approach ”. We were also very pleased to have had Andrea Chegut, Director of the MIT Real Estate Innovation Lab; and Ali Zaidi, Head of Real Assets at LSEG (London Stock Exchange Group) to discuss the use of data science to support real estate investment strategy.
The following are some key takeaways from Prof. Yonder’s recent publication:
- Real estate valuation is complex due to its nature. Every property is unique by its characteristics, location, and neighborhood.
- Although location is fixed, neighborhoods are dynamic and change over time.
- Big data on neighborhoods can help improve real estate investment strategy by:
- (1) better understanding local characteristics and changes in the neighborhoods;
- (2) developing environmental and social benchmarks; and,
- (3) improving prediction accuracy.
- Official language minority, non-permanent residents, external immigrants, are all positively associated with real estate prices for his specific study sample in Montreal.
High-level perspectives from Panel discussion
[Data scientists can find new opportunities for asset managers]
- The real estate sector still lags behind other investment classes in adopting the data science that is research focused.
- It is understood that numerous local factors determine the value but has been difficult brining in that data set. It is now being taken a step further, as to how do you predict the movement of these factors themselves.
- We have so much data; what is missing in our industry is people who know how to do something with it.
- Data can provide targeted and quantitative insights such as how low-income neighborhoods get a greater lift from the presence of trees and street level greenery than high-income neighborhoods.
- We are getting value in learning about what is important for industry via data science.
[Applying machine learning on ESG disclosures]
- Governance is an important predictor of democracy, inclusivity, social well being, and health. The industry is seeing a lack of governance scores in this asset class.
- We would like to see more information related to carbon emissions usage of the real estate assets by the tenants and by the owners.
- Data needs to be standardized, that allows fair comparison from company to company and geography to geography.
Laurentian Bank Professor of Real Estate and Finance at Concordia University - John Molson School of Business
Dr. Andrea Chegut
Director of the MIT Real Estate Innovation Lab
Head of Real Assets at LSEG (London Stock Exchange Group)