New research by WVU statistician Mihyun Kim aims to predict and model extreme patterns in weather events, financial turmoil and pollution levels. Kim notes that her research could play a critical role in West Virginia, where flash floods are common, as seen here in the community of Rainelle after the destructive flash floods of 2016.
(WVU Photo/Pam Pritt)
Using data and numbers, a statistician at West Virginia University is researching new methods to predict and model extreme weather patterns, financial turmoil and environmental pollution.
Kim Mi-hyunAssistant Professor of Statistics at WVU Eberly College of Arts & Sciences School of Mathematics and Data Sciencereceived funding from the National Science Foundation to develop statistical tools to address these problems.
“This research is about more than just numbers,” said Kim. “It‘It’s about using advanced statistical tools to better understand and prepare for extreme weather events that can have profound impacts on our lives and our economy.”
For example, Kim said, her research will answer questions such as: “How likely is it that heavy rain similar to that in location B will occur on the same day in location A?” or “Do the returns of different sectors show similarly extreme daily patterns during a stock market crisis?”
Currently, there is a lack of appropriate statistical tools to analyze extreme behavior in high-resolution curve data, Kim said. Functional data, a set of curves, are observed in various disciplines, such as daily temperature, precipitation, pollutant concentration patterns, and intraday stock return curves. This type of data is not a single value, but rather an entire curve or function, Kim explained.
“For example, instead of just recording how much it rained today, functional data would record how the rainfall changed every minute of the day,” Kim said. “So if we have minute-by-minute data, we can create a smooth line that shows exactly how the rainfall rose and fell throughout the day.”
“Similarly, in finance, instead of just recording the closing price of a stock each day, functional data tracks how the stock price moves throughout the day. This gives us a complete picture of what happened during the day, not just a number at the end. It helps us understand the full story of how things change over time.”
KThe goal of the project is to develop new and innovative methods to accurately model and understand the extreme patterns found in this curve-like data. A key element of this research will be the development of a tool called the “extreme correlation coefficient,” which measures how likely it is that extreme curves in data simultaneously exhibit similar patterns.
She noted that this research could play a critical role in regions with diverse topography, such as West Virginia, where mountainous terrain with streams and rivers is prone to flash floods.
“The 2016 West Virginia flash floods, a once-in-1,000-year event, resulted in 23 fatalities and an estimated $1 billion in damage,” Kim said. “The mountainous topology across the state resulted in varying flood patterns, concentrating risk and damage in certain areas. I believe my risk assessment tools can help develop better preparedness and response strategies, ultimately protecting lives and property.”
The risk assessment tools are also transferable to other mountain states outside of West Virginia and applicable in various disciplines that require risk management, Kim said. She is confident that government agencies will use the tools for disaster preparedness and response.
In addition, their research will help financial institutions manage their risk exposure more effectively.
“Given that millions of Americans have savings in retirement plans, it is imperative to accurately quantify the risk of catastrophic financial loss,” she said. “By providing precise measures of the risks associated with extreme market conditions, this project supports national efforts to maintain economic security.”
The project involves bachelor’s and master’s students in statistics and mathematics research.
TTo increase impact, Kim plans to collaborate with researchers in environmental science, economics and engineering at WVU. She also said she hopes to forge connections with government agencies and organizations to gain access to real-world data and ways to further address these issues.
-WVU-
js/28.08.24
MEDIA CONTACT: Jake Stump
director
WVU Research Communication
304-293-5507; [email protected]
Call 1-855-WVU-NEWS for the latest West Virginia University news and information. WVUT today.
