Research

Research Interests

Statistical and computational challenges of complex data:

• Statistical keywords: Causal inference; High-dimensional statistics; Post-selection inference; Latent variable models
• Applications: Neuroimaging; Genetics; Health studies; Precision medicine; Biomarker evaluation
• Others: Alzheimer's disease; Application of deep learning methods to medical imaging

Publications and Preprints

(* = equal contribution or alphabetical ordering)

• Fighting noise with noise: Causal inference with many candidate instruments.
  Xinyi Zhang, Linbo Wang, Stanislav Volgushev and Dehan Kong. (2024)
  
• Towards the theory of subdata selection for massive data logistic regression with missing responses.
  Xinyi Zhang and Stanislav Volgushev. (2024)
  
• A multiple fixed-sequence procedure for family-wise error rate control with applications to fMRI data.
  Xinyi Zhang, Brian S. Caffo, Martin A. Lindquist, and Zheyu Wang. (2024)
  
• Are there really no returns to education? A re-analysis using post-selection instrumental variable estimation when there are invalid instruments under a multi-group setting.
  Jiaying Gu\(^*\), Harry Krashinsky\(^*\), and Xinyi Zhang\(^*\). (2024)
  
• MRI distance measures as a predictor of subsequent clinical status during the preclinical phase of Alzheimer’s disease.
  Xinyi Zhang, Brian S. Caffo, Anja Soldan, Corinne Pettigrew, Erus Guray, Christos Davatzikos, John C. Morris, Tammie L. S. Benzinger, Sterling C. Johnson, Colin L. Masters, Jurgen Fripp, Susan M. Resnick, Murat Bilgel, Walter A. Kukull, Marilyn S. Albert, and Zheyu Wang. (2024)
• Identifiability of finite mixture models without and with covariates: A nonparametric and semiparametric perspective.
  Xinyi Zhang\(^*\), Zheyu Wang\(^*\), and Xiao-Hua Zhou. (2024)
  
• A non-linear latent variable approach for biomarker dynamics and inference on their temporal order: Application to Alzheimer’s disease.
  Zheyu Wang, Yuxin Zhu, Xinyi Zhang, and Kexin Zhang. (2024)
  
• Joint modeling of multivariate longitudinal biomarkers and disease progression with differentiated covariate effects.
  Zhuojun Tang, Yuxin Zhu, Kexin Zhang, Xinyi Zhang, and Zheyu Wang. (2024)
  
• Supervised principal component regression for functional data with high dimensional predictors.
  Xinyi Zhang, Qiang Sun, and Dehan Kong.
  Journal of Computational and Graphical Statistics (JCGS), 2023 [Link]
  ASA Nonparametric Statistics Section Student Paper Award Finalist, 2018.
• Discussion of “Vintage factor analysis with varimax performs statistical inference” by Rohe & Zeng.
  Xinyi Zhang\(^*\) and Ying Zhou\(^*\).
  Journal of the Royal Statistical Society, Series B (JRSSB), 2023 [Link]
• Considering strategies for SNP selection in genetic and polygenic risk scores.
  Julien St.-Pierre, Xinyi Zhang, Tianyuan Lu, Lai Jiang, Xavier Loffree, Linbo Wang, Sahir Bhatnagar, and Celia M. T. Greenwood.
  Frontiers in Genetics, 2022. [Link]

Working Papers

• A semiparametric method for latent variable modelling with covariate adjustment.
  Xinyi Zhang, Zheyu Wang, and Brian S. Caffo. (2024)
• Assessing the impact of deep learning methods on the cost and need of brain MRI preprocessing for predicting Alzheimer’s disease.
  Shijia Zhang, Xiyu Ding, Xinyi Zhang, Brian S. Caffo, and Zheyu Wang. (2024)