The Science Behind Big Data
From the Spring 2012 OPEN magazine cover story, "The Big Data Machine."
McCombs researchers are at the forefront of Big Data research. Learn about four projects our faculty are working on, from cancer treatment to online advertising.
Search Engine Ads: Location, Location, Location
Ashish Agarwal, Assistant Professor, Information, Risk, and Operations Management
Online advertisers pay extra to be first in the line of ads along the right side of a Google search. But Agarwal found it’s more profitable to be a few slots down. Analyzing the same ad in different positions, he found shoppers are more likely to click on the ad if it sits at the top. But they’re more likely to buy if it appears lower down—in part, because it’s the most recent ad a buyer has viewed.
Stock Message Boards: Don’t Believe Everything You Read
Prabhudev Konana, Professor of Information Management and Director of the Information Management Program
Every day, investors post millions of notes on Internet message boards. Konana studies how those postings sway their decisions to buy, sell and hold. Instead of looking at a stock’s pros and cons, he’s found, investors tend to read messages that reinforce what they already believe. Now he’s investigating whether online sentiments can be used to beat the market. He’s testing statistical models that analyze message board posts to predict the directions of stock prices.
Data Detective: Finding Hidden Causes
Carlos Carvalho, Assistant Professor of Statistics
If there’s a statistician’s version of a Swiss Army knife, it might be Bayesian analysis. A technique for finding patterns in complex systems, Carvalho first used it to pinpoint genes that affect a cancer patient’s chances of recovery. Now, he’s teasing out factors that affect the prices of financial assets, and that increase or reduce risk in investment portfolios. Practical applications could include more secure management of a 401(k).
Better Decisions Through Math
Kumar Muthuraman, Associate Professor of Information, Risk, and Operations Management
What do dialysis and commodity storage have in common? Muthuraman investigates how to use mathematical models and optimization in both fields. He specializes in stochastic control, which addresses decision making in situations where random factors influence outcomes. By optimizing the models, his research can help a doctor decide the best time to switch the location of a dialysis valve, or an investor decide the right prices at which to buy and sell oil.