²ÝÁñ³ÉÈËÉçÇø

Chaegeun Song's research focuses on developing novel statistical methods for analyzing complex, high-dimensional data of various types. Song works at the intersection of dimension reduction, statistical machine learning, and uncertainty quantification, particularly in handling mixed predictors. As data grows larger and more complex, traditional statistical approaches often fail to capture the relationships between variables or become computationally ineffective. This motivates his work on sufficient dimension reduction methods to find a lower-dimensional representation of the predictor variables that retains all the information about the response variable.

Song's teaching centers on empathy, which allows him to connect with his students by recalling his own struggles and triumphs in learning. He aims to foster statistical thinking, and strives to equip students with statistics and data science to think critically about data, make informed decisions, and communicate their findings effectively. Through a data-centered approach, Song creates learning experiences that move beyond textbook problems to real-world applications, emphasizing hands-on experience with real datasets.