Research

Main Research Area

• Statistical Network Analysis

• Log and Process Data Analysis

• Bayesian Adaptive Clinical Trials

Current Ongoing Research Projects

Statistical Network Analysis

Network Mediation Analysis

• We examine how individual traits interact with surrounding social networks using a mediation framework, aiming to disentangle the simultaneous mechanisms of social influence and social selection.

• We extend network mediation analysis by embedding the latent network structure directly into the causal pathway. By incorporating it within a hierarchical data structure, this framework enables the separate identification of mediation effects at both the group level and the overall population level.

Political Data Analysis

• We present a Bayesian framework that integrates Euclidean ideal points with bill-level covariates to quantify how issue content shapes legislators’ voting deviations. This framework yields issue-specific coefficients for every legislator, offering a unified Bayesian approach for studying issue-structured legislative behavior in multiparty systems such as issue-level polarization, intra-party cohesion, and cross-party alignment.

Network Modeling

• We present a Dynamic Latent Space Model (DLSM) framework for analyzing international trade patterns using WITS global value chain data. By modeling temporal network structures, this approach offers an improved alternative to traditional gravity models.

• We combine the Weisfeiler–Lehman kernel values from statistically projected item graphs with latent distances that capture similarity between respondent network positions, providing interpretable insights into homophily and changes in edge states.

• We apply hierarchical structure to a latent space item response model to account for upper-level parameters in response data. The proposed model faithfully reflects heterogeneity among individual respondents, while also effectively capturing complex group cultural structures by explicitly estimating upper-level group parameters.

• • We analyze fMRI-based prospective memory (PM) tasks using Bayesian dynamic network models to identify the neural mechanisms underlying intention processing.

  • We develop a Bayesian hierarchical framework for analyzing fMRI-derived correlation matrices on the SPD (Symmetric Positive Definite) manifold through global-local decomposition. The global pattern is characterized by the Fréchet mean, while subject-specific local variations are modeled as low-rank structures in the tangent space of the manifold.

  • We develop a fully Bayesian zero-inflated factor model for high-dimensional count data with automatic dimension selection. Unlike existing methods, our constraint-aware approach enables scalable and accurate analysis of single-cell RNA-seq datasets.

  • We propose a novel phylogenetic inference framework that compares model-based Bayesian phylogenetics with the geometric perspective of a Hyperbolic Latent Space Joint Model (HLSJM), reconstructing evolutionary trees from inter-language distance structures derived from hyperbolic embeddings.

Log and Process Data Analysis

• We focus on an enhanced item response model that incorporates log data from computer-based testing. The current research involves embedding methods, dimensionality reduction, and statistical analysis of the resulting embedding vectors.

• We focus cross-national differences in behavioral transition structures using problem-solving log data. Multi-state survival models are applied to estimate transition probabilities, which are then embedded into network representations to analyze behavioral similarities and differences across countries.

• We provide a comprehensive tutorial for PIAAC web log data analysis, focusing on three key approaches: feature extraction from action sequences, cognitive process modeling to uncover latent problem-solving strategies, and predictive modeling for performance outcomes.

• We develop nonparametric two-sample hypothesis tests for sequential data by embedding sequences into a signature-kernel-based RKHS, enabling rigorous comparison of complex temporal patterns such as clickstreams.

• We focus on estimating coherent latent positions for each node in a multiplex network, where edges across layers are represented by different types of relational data.

Bayesian Adaptive Clinical Trials

• We develop a novel Phase I clinical trial methodology by accounting for both dose-limiting and low-grade toxicities, refining dose escalation strategies and enhancing patient safety.