Welcome Kengo Kato
- New Faculty Year: 2018
Kengo Kato has recently joined the faculty of Cornell’s Department of Statistical Science as an associate professor. Kato’s research is in the fields of mathematical statistics, econometrics, and economic statistics with a focus on high dimensional statistical and econometric models.
Before coming to Cornell, Kato was an associate professor in the Graduate School of Economics at the University of Tokyo from 2014 to 2018. He was an assistant professor in the Department of Mathematics at Hiroshima University from 2009 to 2013. During his time at Hiroshima, Kato was a visiting scholar in the Department of Economics at the Massachusetts Institute of Technology (MIT) in 2011-2012.
Kato was born in Nagoya, Japan and spent seven years of his childhood in Brussels, Belgium where is father was an executive with the Toyota Corporation. His family returned to Tokyo when Kato was thirteen years old.
When he started his undergraduate studies at the University of Tokyo he chose the law track. “I thought I wanted to be a lawyer,” says Kato. “But during my second year I switched to economics. My real love was mathematics, but at the same time I wanted to be able to apply what I was working on.”
Kato planned to go directly into industry as soon as he earned a Master’s—he even had an offer from a major bank. “During my Master’s program I solved a small problem for my thesis,” says Kato. “The moment of discovery is quite exciting. That experience made me think it might be a good idea to go on and get my Ph.D.” He completed a graduate program in statistics in the School of Economics at the University of Tokyo.
After earning his Ph.D., Kato focused on high dimensional statistics. His research at Cornell aims to extend the central limit theorem (CLT) to modern high-dimensional data sets. An example would be DNA. The human genome has thousands of genes. If you wanted to know which combination of genes might contribute to a specific disease there are a staggering number of combinations to account for. “High-dimensional CLT will be more and more useful, “says Kato, “in analyzing these very large, very complex data.”
Because he has a deep background in economics, Kato is also doing research into economics-based problems. The intersection between economics and statistics is called econometrics. A major concern in econometrics is how to make inferences when there are errors in the data. This is called the measurement error problem (MEP) and Kato is working to address this concern. An example would be how to measure student ability. “There is no way to measure this thing we call “ability” so we use proxy values,” says Kato. “These proxy values have errors built in and we need to correct for those errors.”
Kato is happy to be at Cornell. “There are no Statistics Departments in Japan, so I have always felt a bit isolated professionally,” says Kato. “Cornell has a great history in Statistics. In fact, many of my heroes in stats are somehow affiliated with Cornell.” Kato is looking for graduate students to join in his work. “I am looking for a good match in topics so that a student can learn more about what they want to learn about and at the same time their work can support mine.”
When he is not working on tricky statistical problems, Kato says that his two children, (aged four and six) take his mind away from work fairly quickly.