Eight Shippensburg University students spent the day on Feb. 19 set up in the Math and Computing Technologies Center to compete in a mathematic modeling and science data contest. Members of the math modeling and data science club teamed up in groups of two or three students to solve modeling and data problems in the COMAP weekend-long competition. 

The COMAP (Consortium for Mathematics and Its Applications) contest began Thursday, Feb. 17, at 5 p.m. and ended five days later on Monday, Feb. 21, at 8 p.m. when the answers were to be submitted. The annual contest allows students to use statistical techniques to understand data and formulate answers to open-ended questions. Through a 25-page paper, the groups are to explain their answer and process in a way that would be understood by the general public.

“Doing mathematics in the classroom and in the real world can be quite different,” explained Grant Innerst, an SU professor and the adviser of the club. “Especially with applied mathematics, which is what this would be called. We think it is really important that students get out and get exposed to doing real problems in the real world. That is why we are really excited that students are doing this.” 

To simulate real-world experience, the contest produces a scenario in which made-up companies are hiring competitors to create and explain models. 

This year’s competition is the first that SU students have participated in since 2020 as students chose not to participate in 2021 because of the COVID-19 pandemic. This year, however, three groups of SU students participated — a record number of teams for the math department in recent history. 

The mathematics department at Shippensburg University has a long history of participating in COMAP’s contest, according to Innerst. He even participated in the competition himself when he was an SU student from 2011 to 2015. 

It was the first time participating for all the SU competitors. Each group solved a different type of problem out of the six types that were available for students to choose. 

Junior Katelyn Aumen, senior Megan Woldz and junior Lily Simmons — all mathematics majors — solved problem E. Problem E was a math modeling problem about sustainability. The students looked at carbon sequestration, studying forests and carbon dioxide’s increasing effects on global warming.

They approached their problem by researching the optimal number of trees to absorb carbon dioxide in each area. Through that information, the group began to build a model to find what forest plan is the best for carbon sequestration. 

On the Saturday of the competition, the group’s first two hours were spent on gathering information on past similar concepts to aid them in solving the problem. By that morning, they had taken a few routes that did not pan out, according to the group.

“We kind of hit a roadblock when it came to developing our model,” Simmons said. “But we did have a breakthrough of which variables to put in the equation, so we are not too discouraged.”

Another group made up solely of SU seniors felt confident about where their approach was taking them. Hunter Kindlin, a computer science major, Jerry Marvin, an economics and mathematics dual major, and Steven Hanaway, a mathematics major — all three have data science minors — participated in the event to solve problem C. 

Their problem was a data insights problem that had them use science data to find their answer. The specifics of their problem asked them to compare the market prices of gold versus the market prices of bitcoin daily in order to advise someone on how they should invest each day. 

The group could only use historical data about the market prices for the two currencies. They were only allowed to use this data to aid them to predict the future prices of both. The group was asked to figure out the patterns of the market prices in order to know how someone should invest every day based on the past prices alone. 

Immediately after the start of the competition on Friday, the group got to work to research the problem and start to brainstorm ideas for formula models to test out. By the morning, the individuals in the group had tried many different theories.

“I’ve gone down a lot of paths that have not worked out, but I feel fairly confident with our direction right now,” Marvin said. “We’ve got a lot of different paths that we are trying.” The group had used trial and error on several different models and explained that they were learning as they went.

A side project of Kindlin’s from a few years prior spurred their idea of using the relative strength index (RSI) which is a momentum indicator that is used to measure changes in the price in recent history. Kindlin got the idea from an automated bot that he created that would buy and sell cryptocurrency for him. 

The RSI allowed the group to create a ratio based on if the price increases or decreases. It then puts a heavier weight on the more recent dates in order to emphasize the most relevant price changes. RSI is only relevant to the exact thing it is calculating for, not in comparison to another, the group explained.

In order to remedy this, they calculated a rolling average and rolling standard deviation of the RSI they created in order to look at the past days. They were then able to find a score value in order to properly compare the prices of gold and cryptocurrency. 

The final group was composed of Everest Small, a junior computer engineering major, and Peter de Bruin, a freshman mathematics major. The two addressed problem F, regarding policy. In their problem, they needed to figure out how they could mathematically ensure global equity in the field of asteroid mining. 

The two took to research the current international maritime laws and how they could translate into space asteroid mining. They wanted to find how the different nations can share space.

Their approach is to solve this future problem with current rules and regulations that are in place internationally. On Saturday morning the pair had decided to use the formula of percent global market share divided by percent global gross domestic product.

“This is ending up taking a lot longer than I expected it to,” Small explained. The competition lasting a whole weekend was much longer than the mere hours that Small expected their problem to take them. 

Many of the participants across the three groups explained that they joined the club and entered the COMAP competition because of the passionate advocation professors Innerst and Johnna Barnaby, advisers of the club, shared in their classes. Aumen, Woldz and Simmons, specifically, joined after taking a math modeling course with Barnaby in fall 2021. 

The competition is scored in a series of levels: not judged, disqualified, unsuccessful participant, successful participant, honorable mention, meritorious, finalist and outstanding winner. It is very difficult to receive scores above successful participant or honorable mention, according to Simmons.

In the COMAP contest, there are not necessarily wrong answers, according to Simmons, because contestants are creating an idea. The participants receive a score based on the process and their ability to explain that process in the written paper. 

The Consortium for Mathematics and Its Applications will release the results of the contest on or before May 20, 2022. 

Share