How Zestimate Works
Accuracy is an issue with Zestimates…
I happened to find this article, which explains the processes used to generate Zillow’s Zestimate, the estimate of residential property value: Inside the Zestimate: Data Science at Zillow
I found it interesting that they use the ‘R’ spreadsheet for doing some number-crunching, not Excel. ‘R’ is what George Dell has been demonstrating in his classes. But the graphs are generated by a different software.
Accuracy is an issue with Zestimates, according to Zillow, per the article:
“Nationally, the median error rate is 8.3 percent, the company says.”
However, they don’t define what ‘error rate’ means. Is the Zestimate too high, or too low, by 8%? If a $450,000 home is the target, an 8% error rate difference produces a range between $414,000 and $486,000.
My ‘guestimate’ is appraisers are more accurate than that.
Excerpt from “Inside the Zestimate: Data Science at Zillow” article
If you’re like most homeowners, you probably sneak a peek at your ‘Zestimate’ from time to time to see how your home’s value might have changed….Three times per week, Zillow updates its Zestimate for 110 million homes in the U.S… The company has also gone back in time to create historical Zestimates, which help it to track how property values have changed over time and improve the current Zestimate.
The more data Zillow has on a given home, the more accurate the Zestimate, the company says. The Zestimates are usually accurate to within 10 percent in most American metro areas, Zillow claims. Nationally, the median error rate Zillow_1is 8.3 percent, the company says…
The company is also using GraphLab Create to track month over month (MOM) changes in Zestimates, with an eye for identifying “problematic Zestimates,” McClure writes in his presentation. “We now have a tool that can slice and dice the Zestimate and look at all of our data by any number of factors,” he writes…