A recent CRETech article shared research conducted by Enodo's data science team that identified 9,000 undervalued multifamily properties.Enodo applied its price and cap rate prediction algorithms to its database of 1.7 million assets nationwide to identify the nation’s most statistically undervalued multifamily properties.
According to Enodo’s research, the assets identified by its algorithm are statistically underpriced for their respective markets, and rents in these properties can be raised by as much as 20% with minimal loss of occupancy.
Using machine learning, Enodo developed a method to systematically identify undervalued assets based on their individual performance amongst competing properties. The algorithm flags properties that are either prime value-add candidates or properties that are charging under market rate rents. After identifying the undervalued assets, Enodo applies its machine-generated cap rate to calculate the market value for each asset as it is currently performing, and identifies the potential yield if acquired.
“Enodo currently predicts market rents nationally with under 5.5% median error, and cap rates within 0.35% of actuals. This gives us all the insight we need to determine when an asset is renting below market, and what the incremental income from rent increases will be worth to investors in each market,” said Enodo CEO, Marc Rutzen.
Enodo trained its algorithms on data from more than 21 million apartments, 12,500 operating expense statements, and over 3,000 closed multifamily transactions nationwide to develop its automated valuation model (“AVM”). By predicting rent, operating expenses, and cap rate separately, Enodo is able to accurately replicate the income approach used by multifamily appraisers.
“This advancement by our data science team has created the ability for Enodo to monitor the entirety of the multifamily market and automatically identify opportunities to acquire properties at below market value,” said Rutzen.