The webinar was hosted by Research Analyst Joe Pomerville, who discussed the details and advantages of participating in the SAS Student Shooutout Competition.

The Institute has a partnership with SAS, an analytical software company, and work to organize a student competition to promote application of predictive analytics to solve social, economic and other issues. This year will be the competitions 10th year.

The annual competition challenges students to tackle a complex problem using SAS software and advanced analytics. The student teams are given a collection of datasets and a "problem statement" built and put together by IHBI. The students must use a variety of analytical methods and approaches they see fit to the specific problem. After analyzing the problem, students must come up with a solution and produce a comprehensive report and SAS flows, or codes, to be considered for grading.

"The reports are around 30,000 words," Pomerville said. "The Judges grade on quality and accuracy of reports through several rounds of judging. They usually go back and forth discussing the top three teams to make sure they have the best and most accurate analytical solution. Some honorable mentions may be given as well, depending on topic and quality of papers."

The top three winners will receive complimentary admission o the annual SAS Analytics Conference where they will be awarded for their work and give a presentation on their project. Their universities will also receive financial donations.

Pomerville discusses last year's problem: predicting city wide wind and solar energy production and electrical energy from a hypothetical city. The problem addressed a relevant problem, looking at clean energy usage in the city with real data. There is background information given with the problem and many factors are discussed.

"The students obtain a 'data dictionary' which contains the name and a brief description of each variable," Pomerville said. "For example, the first one would be 'power city weather consumption,' containing the variables that go into the data. There's also things students have to figure out on their own by examining data and finding inconsistencies."

There are with the 14 data sets provided and students can only use those sets. These sets can differ in size, formats and structure, but mostly use SAS files. Sometimes text files and others will be used depending on the year and the problem.

"Some issues that they had to deal with were calculating daylight savings time change when working with wind and solar data," Pomerville said. "Some of the data was adjusted and some wasn't. The teams had to recognize whats alike and different, and also what constituted as missing data. This tests multiple skills, including a formal report to the imaginary city government."

The competition is a semester long, starting with the problem in January. The teams have all semester to work on the problem and provide a solution in June. The problem is put together by IHBI faculty and students throughout the whole year prior.

"The shootout is a great experience for students to learn how to analyze and use data," Senior Researcher and Data Scientist Imad Haidar said. "Although CMU students cannot formally participate due to our connection with the competition, they still have access to the problem for free."

Pomerville shared that the students usually say they really like the shootout because its different from the plain data that they receive in the classroom.

"The problem is start-to-finish," said Pomerville. "This makes it easy for professors to use in classroom and work on all semester. The problems are also very relevant and can reach out to all demographics and subjects. Environmental students and engineering students can benefit from the problem, as well as many more."

At the end of the webinar, students tuning in were able to ask questions to the presenter about the topic, contact information was also provided for the SAS coordinator as well as IHBI faculty.

The webinar was the last of the calendar year, but the schedule for next year will be posted soon.