Last April, two CDS students, Zewei Liu and Olivia Yang, along with two Stern Business School Students, Andrew Hamlet and Troy Manos, won the University of Iowa’s MBA Business Analytics Case Competition. We asked Zewei and Olivia a few questions about their work.
Can you give us a bit of background about the competition?
Olivia: It was a business analytics case competition hosted by the University of Iowa. 14 teams participated and represented universities across the nation. At least two participants from each team had to be MBA students. The competition expected participants to have a strong sense of business and data analytics. The case was provided by a hospital called UnityPoint, who sponsored the competition.
Zewei: We had to give two rounds of presentations to a panel of 14 individuals. The panels were comprised of analytics professionals and faculty from the University of Iowa.
What was the case that your team had to solve?
Zewei: We were given a healthcare case about readmissions. A readmission is when a patient is readmitted to the hospital within 30 days, for the same reason as the initial stay. Readmissions are generally used as a measure for how well a hospital is doing in treating its patients.
The case was provided a week ahead of time, giving us time to do research and data analysis. The deliverables were:
1) to come up with a readmissions risk score for each patient
2) build a guided user interface (GUI) for clinicians to use
3) design a protocol for recommending a post-care plan to a patient
Olivia: The problem was to reduce the readmission rate for the hospital, and provide the optimal post-discharge care plan for the patient, accounting for price and quality of care.
How did you go about solving it?
Olivia: We applied machine learning techniques to predict the readmission rate and see which elements impact the readmission rate for patients. By assigning scores to different post-discharge care plans, we determined which plans have the lowest rate of readmission. Then we built a GUI for patients and physicians, to simplify the process in the future.
You both were on a team with two Stern Business School MBA candidates. Can you talk about working with students from another area of NYU, and what each group of students brought to the team?
Olivia: It was a great experience working with the Stern MBA students. MBA students tend to think differently than most data scientists. Troy and Andrew’s strong business sense enriched the project and made this project run smoothly.
Zewei: Everyone on our team was technically savvy and knew how to code, so working with data and building the interface came somewhat naturally. The hardest part was putting all of the elements together in one coherent story that made sense and actually solved the problem for the hospital. I think that working with a different group of NYU students motivated us to think about the problem from different angles, and forced us to have a comprehensive understanding of the case.
Did any of the other teams have members from data science programs?
Olivia: Yes, most of the teams had at least one student with mathematics/computer science background to balance their skill set.
Zewei: Most of the teams were all MBA students, and some teams had members from analytics programs, and some teams had MBA students in business analytics track programs.
How does the problem you guys had to solve compare to a problem that data scientists in the real world deal with every day?
Zewei: Our case and data was from UnityPoint, and they also have data scientists working to solve this same problem.
Can you talk about how the CDS program prepared you both for the Iowa Business Case Analytics competition?
Zewei: David Rosenberg’s Machine Learning class, Carlos Fernandez-Granda’s Statistical and Mathematical Method’s class, and Foster Provost’s Intro to Data Science class were foundational for me. All the methods that we used in this competition were from the CDS program. But what I found most useful about the CDS experience was how students were encouraged to not just solve a problem, but gain insight into the data. Being able to gain insights from data greatly helped us in this case competition.
Olivia: Machine learning is the key method we applied and the NYU Master of Science in Data Science program does an excellent job of not only teaching us the theory behind data science methods, but also encourages us to gain a sense of real world problems.