Lynna Chu
Position
- Assistant Professor, Department of Statistics
Dr. Chu is always interested in change-point problems. If the datasets are high-dimensional/non-Euclidean in nature, that would be a bonus. But generally, she is interested in applying/developing change-point algorithms to a broad range of settings. Related to this, she is also interested in online detection - meaning that the observations are being generated in “real-time” and changes/anomalies must be detected as the data is generated. This could have applications in disease surveillance or early-warning monitoring systems.
Recently, she has also been developing inference procedures for correlated data structures. For example, she developed a non-parametric test for the presence of random effects in high-dimensional studies. She is also working on non-parametric hypothesis tests for general distributional differences that can properly control type I error in the presence of correlated data.
Recently, she has also been developing inference procedures for correlated data structures. For example, she developed a non-parametric test for the presence of random effects in high-dimensional studies. She is also working on non-parametric hypothesis tests for general distributional differences that can properly control type I error in the presence of correlated data.