By the Numbers

By Edo Steinberg

When Ph.D. student Ryland Sherman heard we want to interview him about his love of big data, he joked that we’ll ask him about how his eyes sparkle whenever someone mentions Excel. It is not uncommon to see Ryland hook up his laptop to the big screen TV in the grad lounge and look through data in as many as 20 open windows all at once.

“I’ve always loved the process of manipulating numbers, and I developed that skillset with a variety of MBA and economics courses I’ve taken,” he says. “Once you get to manipulate numbers that way it becomes kind of a game and a puzzle, and I love playing it.”

“While you can learn so many advanced statistical applications to do really interesting, delving research, in the business world, basic business statistics end up being the primary source for 90% of business decisions,” he says. For that reason, Ryland uses Excel rather than other statistical programs such as SPSS in his research into the economic side of telecommunications. “I learned to use Excel in a day in order to do regressions when I was 20, and from that point on I just love to be in Excel.”

“I love being around researchers, because they always have these great data sets that can often be transferred into Excel for data manipulations and basic statistics,” Ryland says.

He has also noticed that he uses the program much more often than others in the field, and so he has developed skills that complement each other – the ability to conduct research and to quickly create many MBA-style graphs. “I’ve worked with several professors to create graphs and charts for publication and I’m always looking for more opportunities to do that.”

The kind of data Ryland works with is quite different from what most of the department is used to. Rather than conducting surveys, experiments or content analyses, he uses available market data. For example, he and Prof. David Waterman recently had to use available data sets to understand how much the video service Hulu makes in a year from advertising. The information they had included how many hours the average user watches Hulu in a given time period, rough estimates of how many commercials are inserted into a video in an hour, the cost per thousand views of a commercial, the proportion of people who first watch internet videos and then watch Hulu, and the proportion of users who ever watch videos. “We have to make a series of assumptions as to how to put all those things together,” Ryland explains.

“We’re always looking for more data points,” Ryland says. Statical data points make it difficult to show trends across time, so he needs information from different periods. “Every time I find another data point that can contribute to a similar story, piecing something together, it is an exciting thing to say the least.”

Ryland also likes helping others with their data sets. “Ninety percent of the stats that are really needed to lead up to those big statistical analyses are just basic stats. The problem is whether you can rapidly select the data that you need to create all the counts, figures and numbers for other conditional analyses. It is at that stage that I am most helpful.”

Previous Post
Leave a comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: