INDIANAPOLIS, Ind-- Selling a home can sometimes be a headache. Depending on the situation, sometimes months or even years can go by before a buyer is found.
A professor at IUPUI believes he can take the guess work out of knowing when a particular home will find a buyer.
Associate Professor Mohammad Al Hasan says he’s spent the last year studying more than 10,000 home sales in central Indiana. He asserts that by examining the variables of each sale (bedrooms, bathrooms, square footage, neighborhood, etc.) that he could come up with a predictive model for future home sales.
“This is something that can be useful to people, but hasn’t been provided in any of the existing websites,” Hasan said.
The “websites” Hasan is referring too are popular home buying sites such as Trulia or Zillow. Hasan says he incorporated information from those sites into his data when coming up with his algorithm.
“So that information was vital for me to make these prediction systems,” he said.
Hasan’s algorithm methodology is based off of another predictive model designed to predict survivability rates in hospital patients. His version of the model is, of course, altered to reflect the real estate market.
“So we bring that idea of survival analysis, the methodology more or less, to this problem of home sale prediction. In this case we’re not interested in survival, we are interested in the probability of when a home will be sold,” he said.
Essentially his model replaces survivability with marketability.
Hasan says his formula could also provide home sellers with information on how certain changes to the home could affect its sale timeline.
He says there is still some work to be done to the algorithm, including adding variables that account for time of year the home is put up for sale. He says a few home sales companies have already approached him to make his algorithm commercially available. He says that will likely happen sometime in the future.