Media Coverage

IN HIS book, “Capital in the Twenty-First Century”, Thomas Piketty argues that it is impossible to find an “objective basis” for the high salaries of senior executives in terms of their individual productivity: they pay themselves such exorbitant sums simply because they can. However, in a forthcoming paper in Management Science, an American journal, two academics claim to have found such an objective measure, and conclude that most bosses are not overpaid.

In their study, Bang Dang Nguyen of the University of Cambridge’s Judge Business School and Kasper Meisner Nielsen of the Hong Kong University of Science and Technology looked at how firms’ shares react when the chief executive or another prominent manager dies suddenly. They identified 149 cases of this happening at American companies between 1991 and 2008.

Consumers have become highly adept at avoiding television advertisements. We switch channels, divert attention to our tablets and phones, and of course fast-forward through ads on our DVRs. Partly in response to this loss of attention, marketers are increasingly focused on product placement as an alternative way of exposing us to their brands. After all, product placement is innately much harder to skip given its integration into the actual program content.

Most academic research on product placement has primarily considered it as a separate persuasive technique independent from the commercial break advertising. That is, mirroring early research on TV advertising, research has focused on how product placement influences viewers’ recall of and attitude toward brands. However, this overlooks the possibility that product placement in a show might influence the likelihood of viewers watching an advertisement for related products at the next commercial break.

In research just published in Marketing Science, my colleagues David Schweidel of Emory University and Natasha Foutz of the University of Virginia and I began to explore whether such synergies exist.

Much like the definition of big data, the job description for data scientist is definitely a work in progress. What skills are required? Well, in addition to possessing a strong math and computer science background, including the ability to devise algorithmic solutions to complex problems, data scientists need to be good communicators -- people capable of grasping business issues and explaining data-driven insights to executives and managers.

Fair enough. But how does an organization know that the data scientist it's just hired has all of these skills? The Institute for Operations Research and the Management Sciences (INFORMS), a 10,000-member international organization of analytics professionals, believes its recently launched Certified Analytics Professional (CAP) program provides the rigorous certification needed in the burgeoning big data field.

"We realized that there were a lot of people putting themselves out there as analytics practitioners, but no real defining body on what that meant," INFORMS president Anne Robinson told InformationWeek in a phone interview.

[Adam] McElhinney said he thinks Enova's hiring practices and goals are what you'd find at most leading-edge analytics shops. So here are eight things you need to know if you're aiming to pass McElhinney's or his ilk's scrutiny and land a coveted analytics job...

2. Get involved with a professional society. The emphasis here is on involvement. Don't just show up for meetings but become involved in some way -- offer up your web development skills if you have them, or help write a newsletter or volunteer to help plan an event, McElhinney said.

3. Show continued development. McElhinney suggested three ways to make your resume stand out in this respect. One, get an analytics-related certification, be it via an independent organization like INFORMS, platform-specific from a vendor like SAS, or related to a specific domain, like finance, risk, or actuarial science. Two, enter a data analysis competition at Kaggle or the like. You don't have to win, but do be prepared to talk about the experience. Three, participate in open-source software development, maybe creating a package for R, working on a Python data analysis tool, or showcasing projects on GitHub.