Students reimagine rancher's 30-year-old cattle market prediction model

A group of students at the South Dakota School of Mines and Technology (SDSMT) have developed software to help farmers better predict the cattle market.

The three students – Jordan Baumeister, Dustin Reff, and Trevor Borman – are computer science majors that used artificial intelligence and data science to create models predicting future market trends. These models also provide a comparison for anomalies, like droughts or floods, using historical data trends.

"Our overall goal was to optimize the risk versus reward tradeoff that shows up when you exchange these contracts on the futures market," Reff says.

The idea to create a model like this came from Ron Ragsdale, a rancher who worked with SDSMT student Todd Gange in 1993. Ragsdale had created a system for calculating predictions of the cattle market on pen and paper, using a series of equations.

"What he did was genius," Gagne says. "He looked at the futures market for both cattle and corn and backed out all the costs needed to fatten his calves. He used 187 variables, not just feed. He included the costs of the lights in his barn, vaccination, and fuel. This way, he knew what he could pay for his calves to make a profit in the future."

Gange helped Ragsdale optimize his system using a computer program Gange and his wife Holly had created in college. Ragsdale's method, combined with Gange's program, failed to produce accurate results only twice.

Ragsdale passed away in 2021 before he was able to publish his thesis on market theory. Before his death, Gange and Ragsdale launched a student program, and Gange passed his software on to a new team of SDSMT students.

The goal of this new team – Baumeister, Reff, and Borman – was to take modern tools like AI and enhance the old program, making it more robust and better able to predict factors that may drive the market off its typical course.

"If I know what the value should be in the future, what happens when something like mad cow disease, widespread drought, or widespread flooding occurs? All these things can send the market into arbitrage," Gagne says. "We twisted and turned this data and tried to look at it in new ways to see anomalies or patterns that we think are tradable in the future."

The team created two prediction models: one that evaluates the historical data trends to determine risk vs. reward and one that predicts the best times to buy and sell.

The project is still ongoing. The three students have graduated but will coach the next team in the fall to begin the next phase of the project. Next semester's students will rebuild the models and determine which of Ragsdale's 187 variables are the most important in predicting the markets and build in indicators for over- or undervalued animals.

"As a sponsor, Todd provided years of data, support, and a good story," says Brian Butterfield, a lecturer of computer science and engineering at Mines. "These students took advantage of the opportunity by applying their skills in data science and data analysis to advance the work. I appreciate watching what emerges by providing students with the framework to build something and gain real-world experience."