Deborah Okoli, a Nigerian PhD student in Applied Mathematics at Mississippi State University, has developed a machine learning model designed to make online market forecasts clearer and easier to understand.
Speaking about her research, she explained that the project aims to help businesses and policymakers base decisions on insights they can “question and trust.”
“My goal is to build machine learning models that don’t just give answers, but explain how they got there,” Okoli said.
Her method, which she calls lag-aware machine learning, studies not only what influences e-commerce growth but also when the effects take place. Factors such as labour productivity, R&D, sales, employment, and investment are examined for both their immediate and delayed impacts.
Okoli uses tools like feature attribution and partial dependence plots to show how each variable contributes to forecasts. She also emphasised rigorous testing: “I use rolling-window cross-validation and stability checks to make sure the models are not missing patterns.”
The practical outcome, she said, is simple but powerful: “A spike in productivity might mean increased online sales in the next quarter, while the effects of R&D spending may unfold more gradually. These insights empower businesses to plan inventory, adjust logistics, and make proactive digital strategy decisions.”
Looking ahead, Okoli hopes to make the system widely accessible.
“I want to create plain-language forecasting tools for non-technical users and develop AI templates that small businesses and agencies can adopt,” she said.
She stressed that clarity is central to her work: “Forecasts should come with seatbelts. If the outcome could change drastically due to one new data point, decision-makers need to know that up front.”
2 comments