Projecting Chemical Feedstock Forecast Outcomes to Assess Impact to Ecolab
Program: Data Science Master's Degree
Host Company: Ecolab
Location: Not Specified (remote)
Student: Sam Barr
Ecolab is the global leader in water, hygiene, and infection prevention solutions and services that protect people and vital resources. Many of their core products are chemicals, and to produce them, they need to procure raw materials from the marketplace. These raw materials are often influenced by the movement of chemical feedstocks, also called commodities. Ecolab tracks the market prices of commodities through third-party data providers. Not only does this help to measure how market movements are impacting the business, but it also allows employees to be proactive and anticipate future price movements. As such, the ability to accurately forecast commodity prices would be hugely beneficial to the company. At the time of this project, Ecolab had simply used forecasts provided by third-party data providers as given, and analysis had not been performed to assess the accuracy of these forecasts. The goal of this project was to assess historical accuracy and attempt to improve upon the provided forecasts. The scope of the project included a select set of commodities based on Ecolab’s exposure to them. This project assessed historical forecasts, produced confidence intervals based on historical performance, generated a scaling factor for which to adjust future forecasts, and leveraged macroeconomic indicators to potentially improve forecast performance in the future.