Slashdot Mirror
DePaul University To Offer Degree In Predictive Analysis
itwbennett writes "The Chicago-based DePaul University will offer what it says is the nation's first master's degree in predictive analysis, the school announced on Wednesday in conjunction with IBM, which will provide resources for the program. 'We realized there was a need to create a program that prepared students in careers in data analytics and business intelligence,' said Raffaella Settimi, an associate professor at DePaul's College of Computing and Digital Media, who helped craft the program. 'A lot of the professionals who work in these fields have a variety of backgrounds, but there really isn't a program dedicated to data analytics,' Settimi said."
There are, however, many quality degree programs in Statistics. As someone who went through one of them, you can largely choose your own mix of theory and practice. I wonder if this isn't just statistics rebranded? I hope it doesn't concentrate too much on certain proprietary software packages. Statistics is like anything else. You can easily produce a bunch of numbers and compile massive books of tables and graphics. But if you don't know the assumptions of each of your methods, and consequently their shortcomings in each situation, you can draw some fairly bad conclusions rather quickly. I just hope this program gives a solid background in theoretical statistical inference, experimental design, and regression analysis, so students understand the 'why'.
I have an M.S. in predictive analytics, and I'm only months away from my Ph.D. These guys didn't do much research.
Get your Astrology and Fortune telling jokes out of the way, but it's not really hard to use predictive analysis and come out to be very accurate (within 3%). You are not always going to hit it on the head, but it's very easy to get close. There are patterns in everything people do and while there will always be outliers for the most part behavior follows basic patterns or cycles, although these may change over time.
It's important to be predictive in anything that involves a queuing system, such as a sales website or where a lot of the theory got its start in telephone calls. Really it applies to anything where people try to do the same thing at once that requires resources for the company to allocate. It can be the difference between success (Blizzcon ticket queue never crashed out) and failure (iPhone 4G). Especially with more people trying to access the same thing at once on the internet, this sort of analysis of how many resources to allocate efficiently is going to be more important. You don't want to allocate too much due to costs, but you can't do too little otherwise your customers suffer.
Yeah, it’s a lot like statistics, but it's going from 'What Happened?' to 'How and why did it happen?' and from 'What is happening now?' to 'What's the next best action?'.
Are the guys teaching this stuff the same ones that failed to predict the financial bubble?
Seastead this.