Department of Economics

2010 Data Mining

SMU Captures Silver Medal in the 2010 SAS Data Mining

A team of Ph.D. students in the Industrial Engineering Department at Oklahoma State University edged out the SMU Data Mining team in the 2010 SAS Data Mining Shootout competition.  The order of finish of the competition was announced on Monday, October 25 at the SAS Data Mining Conference held at Caesar’s Palace in Las Vegas, Nevada.   The SMU team made up of Subhojit Das, Greg Johnson, and Jacob Williamson finished second in the competition.  Das and Johnson are third year students in the Economics Ph.D. program and Williamson is a second year Applied MA student in the Economics Department.   A team of Ph.D. students from the Information Systems and Quantitative Sciences (ISQS) Department at Texas Tech University finished third.  

The problem statement of the 2010 SAS Data Mining competition was posted to all U.S. universities in February of this year.  All solutions were due by July 19, 2010.  Proposed solutions involved analyzing the medical, demographic, and behavioral data of 50,788 individuals some of whom had diabetes.  The task was to determine the economic benefit of reducing the Body Mass Indices (BMIs) of a selected number of individuals by 10% and to determine the cost savings that would accrue to the Federal Government’s Medicare and Medicaid programs as well as to the economy as a whole from the implementation of the proposed BMI reduction program.   Sixty applications were received from Colleges and Universities across the nation with 17 of the 60 qualifying for final submission.  The completion is judged on the applicability and accuracy of the proposed solution algorithm and the quality of the submitted report detailing the results produced by the solution algorithm.      

This is the third year in a row that SMU’s Economics Department has fielded one of the top three Data Mining teams in the nation in the annual SAS Data Mining Shootout competition.  The previous two SMU teams finished as National Champions.   Fomby is already in the process of putting together another team to participate in the 2011 SAS Data Mining Shootout.  Winners of that competition will be announced at the 2011 SAS Analytics conference to be held in Orlando, Florida in October of 2011. 


Left to right: Tim Rey (Dow Chemical), Subhojit Das, Tom Fomby, Greg Johnson, Jacob Williamson, Tracy Hewitt (Institute for Health and Business Insight, Central Michigan University)

  

 

One of the Top 3 Data Mining Teams

The SAS Institute in Cary, NC has just announced that SMU’s Data Mining team is one of the top three finishers in the Nation for the 2010 SAS Data Mining Shootout co-sponsored by SAS, Dow Chemical, and Central Michigan University Research Corporation. The order of finish will be announced at the 13th annual SAS Data Mining Conference Tuesday morning, October 26, 2010 at Caesar’s Palace in Las Vegas, Nevada. The three students on the team, Subhojit Das, Greg Johnson, and Jacob Williamson, along with their faculty sponsor, Tom Fomby, will receive all-expense-paid trips to Las Vegas to attend the conference and the award ceremony. Misters Das and Johnson are third-year students in the Economics Ph.D. program while Mr. Williamson is in the last semester of the MA in Applied Economics program.

The problem statement of the 2010 SAS Data Mining competition was posted to all U.S. universities in February of this year. All solutions were due by July 19, 2010. Proposed solutions involved analyzing the medical, demographic, and behavioral data of 50,788 individuals some of whom had diabetes in order to determine the economic benefit of reducing the Body Mass Indices (BMIs) of a selected number of individuals by 10%. To maximize the benefit of the BMI reduction program, individuals have to be carefully selected for the program. Accurate selection of “high risk” individuals for treatment is quite important because the more accurate the selection of appropriate individuals, the greater the savings on future medical expenses relative to the costs of reducing the BMIs of a limited number of individuals. Evidently SMU’s Data Mining Team proposed a very strong solution offering one of the highest benefit-to-cost ratios for the program among the proposed solutions of approximately 50 teams from universities and colleges across the nation.

This is the third year in a row that SMU’s Economics Department has fielded one of the top three Data Mining teams in the nation in the annual SAS Data Mining Shootout competition. The previous two SMU teams finished as National Champions.

http://www.smu.edu/News/2010/data-mining-30aug2010.aspx