Research

In recent years, my research has expanded rapidly, focusing on the following themes.

Modeling and inference for high-dimensional time series (HDTS): The topics of interest include a broad range of classical and established techniques in time series modeling, such as sparse modeling, dimension reduction, factor models, clustering, graphical models, spectral analysis, change point detection, nonlinear models, and continuous-time models. These methodologies have been extended to high-dimensional settings, with a particular emphasis on incorporating network structures. Ongoing research focuses on HDTS modeling through sufficient dimension reduction, statistical inference for nonlinear HDTS models, and clustering of HDTS-based networks.

Computational and theoretical aspects of statistical machine learning: The topics of interest are motivated by the challenges of fitting high-dimensional models, where efficient and scalable algorithms are essential. Through the lens of high-dimensional statistics, I am also interested in the theoretical foundations of these methods, including the validation of their statistical properties and computational efficiency. Ongoing research focuses on coordinate descent algorithms for nonconvex optimization, empirical risk minimization for dimension reduction techniques, and high-dimensional modeling of functional data.

Methodological applications in scientific domains: The topics of interest are motivated by the development of methodologies tailored to problems arising from specific scientific domains, even when the underlying models and algorithms are relatively simple. This perspective is rooted in my belief that statistics cannot exist in isolation. My research applies statistical tools motivated by problems in psychology, neuroscience, psychiatry, finance, and economics to provide practically useful solutions. Ongoing research focuses on network analysis of multimodal neuroimaging data and the analysis of behavioral and neuroimaging data using high-dimensional nonlinear regression methods.

I am always happy to discuss research ideas and opportunities for collaboration with students, researchers, and anyone interested in related themes.

Publications

• Kim, Y.*, Deb, N.*, and Basu, S. (2025). cxreg: An R Package for Complex-Valued Lasso and Graphical Lasso. To appear at Journal of Open Statistical Software.

• Qiu, Y., Carter, E., Benda, N., Sirey, J. A., Kim, S., Kim, Y., Yu, Z., Kiosses, D. N., Marino, P., Alexopoulos, G. S., Faith, G.-D., and Banerjee, S. (2025). Improving Adherence to Psychotherapy and Clinical Outcome in Patients with Late-Life Depression through Gamified mHealth Technology. American Journal of Geriatric Psychiatry, 34(4), 568–579.

• Kim, Y., Dücker, M. C., Fisher, Z. F., and Pipiras, V. (2026). Latent Gaussian Dynamic Factor Modeling and Forecasting for Multivariate Count Time Series. Journal of Time Series Analysis, 47(1), 43–58.

• Kim, Y., Basu, S., and Banerjee, S. (2025). A Co-Segmentation Algorithm to Predict Emotional Stress from Passively Sensed mHealth Data. Statistics in Medicine, 44(10–12), e70099.

• Kim, Y., Fisher, Z. F., and Pipiras, V. (2024). Group Integrative Dynamic Factor Models with Application to Multiple Subject Brain Connectivity. Biometrical Journal, 66(8), e202300370.

• Fisher, Z. F., Kim, Y., Pipiras, V., Crawford, C., Petrie, D., Hunter, M., and Geier, C. (2024). Structured Penalization of Heterogeneous Time Series. Multivariate Behavioral Research, 59(6), 1270–1289.

• Fisher, Z. F., Kim, Y., Fredrickson, B. L., and Pipiras, V. (2022). Penalized Estimation and Forecasting of Multiple Subject Intensive Longitudinal Data. Psychometrika, 1–29.

• Ahn, D., Kim, K.-K., and Kim, Y. (2020). Small-Time Smile for the Multifactor Volatility Heston Model. Journal of Applied Probability, 57(4), 1070–1087.

Preprints

• Crawford, C., Dücker, M. C., Kim, Y., Pipiras, V., and Fisher, Z. F. (2026). Analytic Standard Errors for Latent Gaussian Discrete-Valued Multivariate Time Series. [arXiv] 

• Deb, N.*, Kim, Y.*, and Basu, S. (2026). Inference for High-Dimensional Sparse Spectral Precision Matrices. [arXiv] 

• Bowling, C. B., Levitan, E. B., Ringel, J. B., Kim, Y., Enogela, E. M., Goyal, P., Glover, L., Reid, R.-J., Sterling, M. R., Safford, M. M., and Banerjee, S. (2026). Characterizing Reserve Across Multiple Domains among Older US Adults: Findings from the REGARDS Study.

• Kim, Y., and Baek, C. (2026). Latent Community Paths in VAR-Type Models via Dynamic Directed Spectral Co-Clustering. [arXiv] 

• Kim, Y., Basu, S., and Banerjee, S. (2025). Identification of Emotionally Stressful Periods Through Tracking Changes in Statistical Features of mHealth Data. [arXiv] 

• Kim, Y., Fisher, Z. F., and Pipiras, V. (2025). Joint Modeling and Inference of Multiple-Subject High-Dimensional Sparse Vector Autoregressive Models. [arXiv] 

• Kim, Y., Loh, P.-L., and Basu, S. (2025). Exact Coordinate Descent for High-Dimensional Regularized Huber Regression. [arXiv] 

A current list of my publications is available on my Google Scholar profile.