Machine Learning & Real Estate
Real estate valuations are unique when compared to other assets, as they- on average- change hands less often, yielding to much more question in the real value. In the gray area between transactions, theoretical valuations are what we must rely on, an amount that is heavily driven by the "cycle," we estimate that we are in the midst of.
However, this is much easier said than done.
The harsh reality is that to predict rent, many valuables must be known: regarding construction, as well as, vacancy rates.
Beyond this, one must also have a vast knowledge of demand and supply models, coupled with monetary policy. Even so, there can be a gamut of factors that can throw all the intelligence that we hold to be self-evident to be overturned entirely.
One thing that we know about real estate prices is that they tend to be cyclical no matter the market we are examining.
However, what can vary across markets is the length and magnitude of these cycles. Given this, many points are brought into question, all of which can be monitored and solved through machine learning. I realize that this article that I am referencing is only one perspective and that the possibilities are endless, as the author acknowledges.
Furthermore, what makes this conversation even more impressive is the impact that COVID-19 has had on the current climate in the real estate market.
After reading an article by Forbes entitled "Real Estate Will Never Be The Same Post Covid-19. Three Top Developers Explain the New Normal", my mind was opened even more so to just how definitive this current situation is!
There are many people, especially the more fortunate, who may be more concerned about the size or location of the real estate, instead of other who are just grateful to have a roof over their head. Interestingly enough, this crisis has redefined how we see our homes.
Our homes have become our whole life, and we are forced to reflect on how lucky we are to have all that we do. Real estate has a "monopoly hold" on human necessity, we can't devise a society without a need for real estate. As Winston Churchill said, “Land monopoly is not only monopoly, but it is by far the greatest of monopolies; it is a perpetual monopoly, and it is the mother of all other forms of monopoly.”
This is something that we can only hope continues in post-quarantine life. Given this, it seems that we will see a complete overhaul in the real estate development market in how we build and what we value. The concepts of being "bigger, wider, and flashier" have become more out of date, and we are more focused on the community.
The question is, how will this impact development after we are out of this pandemic?
We are reminded every day just how much we must take into account the future, and this is no different for developers.
Often, developers have been doing the same things for years, and it has worked. If it's not broken, why fix it? The reality is that it is broken, and they must figure out what it is they will do next. Will health and wellness trump luxury?
Will the community be valued over being tucked in private enclaves with extra security?
Only time will tell. However, we must be careful not to overcorrect and to take every trend with a grain of salt. After all, a trend is simply a fashion.
Written by Jack Argiro
Edited by Alexander Fleiss
Constantinescu, Mihnea. “Machine-Learning Real Estate Valuation: Not Only a Data Affair.” Medium, Towards Data Science, 12 Feb. 2020, towardsdatascience.com/machine-learning-real-estate-valuation-not-only-a-data-affair-99d36c92d263.
Taylor, Peter Lane. “Real Estate Will Never Be The Same Post Covid-19. Three Top Developers Explain The New Normal.” Forbes, Forbes Magazine, 24 Apr. 2020,