Forecasting retail sales via satellite

From Mark Perry’s blog Carpe Diem:

“As part of a growing trend among hedge funds and Wall Street firms, Cold War-style satellite surveillance is being used to gather market-moving information.  The surveillance pictures are often provided by private- sector companies like DigitalGlobe in Colorado and GeoEye in Virginia, which build and launch satellites and take pictures for US government intelligence agency clients and private-sector satellite analysis firms.

That means there are two links in the chain before the satellite data gets to Wall Street—a satellite firm takes the pictures and sells them to an analysis firm, which scrutinizes the images and sells the aggregated data to hedge funds and Wall Street analysts.

UBS analyst Neil Currie had been looking at satellite data on Wal-Mart during each month of 2010, and he’d concluded that there was enough correlation between what he was seeing in the satellite pictures of Wal-Mart’s parking lots to the big-box chain’s quarterly earnings, that he was ready to incorporate that data into UBS’ report on Wal-Mart, which releases its earnings on Tuesday.

Currie purchased his analysis from a small two-year old Chicago-based firm called Remote Sensing Metrics LLC, which had scoured satellite images of more than 100 Wal-Mart stores chosen as a representative sample. By counting the cars in Wal-Mart’s parking lots month in and month out, Remote Sensing Metrics analysts were able to get a fix on the company’s customer flow. From there, they worked up a mathematical regression to come up with a prediction of the company’s quarterly revenue each month.

UBS predicts that Wal-Mart’s second quarter sales will be up from the first quarter, but down a percent against the same period a year ago. But the satellite analysts figure that the number will come in 0.7 percent higher—not lower—based on the traffic surge they saw in the parking lots.”

Read more here.

Print Friendly, PDF & Email

Leave a Reply

Your email address will not be published.