Mining students design software to predict cattle market with 29-year-old program dreamed up by Rancher

RAPID CITY, SD (June 14, 2022) – For many years, investors on Wall Street have used sophisticated software like artificial neural networks to gain trading advantage. These software tools use a range of data inputs and historical trends to predict stock prices.

But the cattle market is a different beast. “The software tools used to predict the stock market fail miserably if you apply them to livestock futures markets,” says Jordan Baumeister. She worked last year with fellow computer science majors Trevor Borman and Dustin Reff to build models that could better predict livestock and corn markets in a bid to give commodity traders an edge. The team used artificial intelligence and data science to create mathematical models to predict future market trends and provide comparison for anomalies, such as droughts or floods, using analytics historical data.

“Our overall goal was to optimize the risk/reward trade-off that arises when you trade these contracts in the futures market,” says Reff.



To achieve this goal, the students had to build on decades of previous work.

A long history of success



In 1993, Todd Gagné was a student at Mines and developing his own software when he crossed paths with Ron Ragsdale, who ran a ranch on 55,000 acres of rolling prairie near the confluence of the Belle Fourche and Cheyenne rivers.

Ragsdale came to breeding after a distinguished career in law as well as a background in mathematics and statistics. He developed his own system for predicting the cattle market using a series of equations he developed by hand with pencil and paper. The model helped him determine when to buy and sell corn and livestock. The two products are linked because cattle are often fattened with maize.

“What he did was kind of genius,” Gagné says. “He looked at the cattle and corn futures market and wrote off all the costs to fatten his calves. He used 187 variables, not just power. He included the costs of lights in his barn, vaccinations, fuel, everything. That way, he knew what he could pay for his calves to make a profit in the future. If the model showed Ragsdale that he could not profit that year, he would lease his land to other ranchers.

Ragsdale asked Gagne to help him improve his equations with a computer program he and his wife, Holly, created when they were college students in 1993. He considered everything statistics and math. He taught me a lot and was an excellent mentor,” Gagné said.

The software they developed was used successfully by Ragsdale for decades to come. He only managed to predict positive results twice, once on September 11, 2001, the other the recession of 2008 with the collapse of Lehman Brothers. “Everything else, the model held on. It bent, but it didn’t break,” explains Gagné.

Gagné graduated from Mines and pursued a career in software development. Today, he is an entrepreneur-in-residence at the university. He acts as a consultant to start-ups. But he never completely lost touch with Ragsdale. The two remained friends over the years and continued to work on the project, adjusting the schedule and learning as they went. Ragsdale ended up writing a lengthy unpublished thesis on his market theory before passing away in 2021 at age 72. Before he died, he worked with Gagné to launch the student project.

“It’s an intellectual curiosity that started as a stampede and evolved into something much bigger,” says Gagné.

To buckle the buckle

In the fall of 2021, Gagné shared the software he and his wife Holly developed as university students, nearly 30 years before, with a new team of Mining students. Gagne sponsored the team’s work and challenged them to use modern tools like artificial intelligence and data analytics to delve into decades of livestock market data and improve the program. original.

The goal was to make the software more robust to better predict commodity prices when outside factors cause the market to deviate from its normal 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 happens, all of those things can send the market into arbitrage “says Gagne. He asked the students to create software that could better predict what to do when the market gets goofy.

“We tweaked and adjusted that data and tried to look at it in new ways to see anomalies or patterns that we think are tradable in the future.”

The student team spent an entire year working on the project. “The computational complexity was huge,” says Baumeister.

The team overcame challenges such as filtering out the noise in the data to get to the heart of the information needed to predict the markets and to focus on the key variables that have the most impact on the commodity process. They ran their model using historical numbers and worked through many iterations of the program until it could predict the known outcome as accurately as possible.

By the end of the year, the team developed two different computer models to help improve commodity trading. One looks at historical trends to help determine risk versus reward analysis. The other, a predictive model, calculates the best times to buy and sell. “We developed a tool to help play the commodity trading game a little better and gain an edge over the competition,” says Baumeister.

The project is ongoing. Baumeister, Reff, and Borman have all graduated and started their careers as computer engineers, but will brief a new team of students in the fall of 2022 to help launch the next phase of the project. “I was very happy, these students are all doing different jobs, but they’re ready to come back and help the next team get to the next phase,” Gagné said.

Over the next semester, the new team will rebuild the model, then work on a sensitivity study to understand which of the 187 variables has the most weight in the model. They will run more historical market data into the model to see how it behaves over time, and they will incorporate indicators as the model runs that will check when the animal might be overvalued or undervalued. -rated.

Mines faculty members overseeing graduate computer engineering research projects are pleased with the progress. “As a sponsor, Todd has provided years of data, support and a good story,” says Brian Butterfield, senior lecturer in computer science and engineering at Mines. “These students took the opportunity to apply their skills in data science and data analysis to advance the work. I enjoy watching what emerges providing students with the framework to build something and gain real world experience. »

A group of Mining computer science majors Jordan Baumeister, Dustin Reff and Trevor Borman have teamed up with a college alumnus to create new software that helps predict the cattle futures market. SDSMT
courtesy picture

Abdul J. Gaspar