Skip to main content
News

Brewing up a Statistical Storm

Bart_Hartman-1080x720.jpg
Photo by Rob Johnson

After graduating in 2006, Brian Hartman is back at BYU, not as a student, but as director of the actuarial program and a new assistant professor in the Department of Statistics. He brings with him his wife, three kids, and a PhD in statistics from Texas A&M University.

“I’m excited to teach BYU students,” Hartman said.  “BYU students are fun. I feel like you can make an impact in a lot of ways here at BYU.”

Outside of teaching, Hartman has applied statistics to a number of industries such as health and property insurance, international marketing, finance, and law.

“I love being able to help people solve a variety of problems,” Hartman said. “There’s a famous quote, ‘statisticians get to play in everyone’s backyard’. . . I get to learn about different things that I never thought I would learn.”

Perhaps the most memorable work Hartman has taken a part in was working with utility companies in New England to predict the amount of damages that severe weather could cause.

“It gives the public information, and it gives the utility company information,” Hartman said. “If there’s a big storm they can’t handle with just their crews . . . they can effectively hire outside crews to come in and help repair the damage.”

Hartman’s first major storm damage prediction was in October 2012, for Hurricane Sandy. His team received the data the week before and hadn’t had time to properly test any models.

“The storm is coming, it’s starting to rain,” Hartman said. “We know the power’s going out pretty soon. I’m working on my desktop computer . . . and we’re trying to fit a new model.”

Running short on time, Hartman fit a very simple model (just damage as a linear function of wind) and predicted that the storm would cause 13,300 damages to trees, houses, and property. When the actual results came in, the utility company was amazed at how close Hartman was with his prediction. The storm had caused 14,100 damages.

“We never told them that it was me in my apartment on a computer, [thinking,] ‘Let’s pick the simplest model we can,’” Hartman said. “The model we have now is much better, and much more sophisticated, and [it has been] tested and confirmed. But that first model was a complete guess!”

These experiences have not only allowed Hartman’s knowledge in statistics to grow, but also in meteorology.

“I know more about weather forecasting now than I ever thought I would in my whole life, just because of this utilities project,” Hartman said.

Hartman received his Bachelor of Science in Actuarial Science from BYU in 2006, and Ph.D. in Statistics from Texas A&M University in 2010. Following graduation, he received a post-doctoral fellowship in Actuarial Science from the University of Waterloo. Prior to joining the BYU Department of Statistics, Hartman was an assistant professor of Actuarial Science at the University of Connecticut.