Research

✝ indicates that I am the corresponding author. * indicates students under my supervision/co-supervision. # represents equal contribution.

Preprints

1. Yiling Kuang *, Xinyuan Song ✝ and Kai Kang ✝ (2025+). A Multi-Label Growth Curve Change-point Model for Characterizing Multifaceted Dynamics in Multivariate Longitudinal Data. (Submitted)

2. Zijian Ye *, Yuanjia Wang, Yu Gu and Kai Kang ✝ (2025+). A Longitudinal Multi-Domain Latent Structure Model for Domain-specific Disease Progression. (Submitted)

3. Junxuan Chen *, Xiangnan Feng and Kai Kang ✝ (2025+). Latent Gaussian Process Joint Model for Integrative Analysis of Multimodal Biomarkers and Initiation of Medication of Parkinson’s Disease. (Submitted)

4. Kai Kang ✝ and Yuqi Gu ✝ (2025+) A Blockwise Mixed Membership Model for Multivariate Longitudinal Data: Discovering Clinical Heterogeneity and Identifying Parkinson's Disease Subtypes. (Submitted) [arXiv]

5. Yuyang He *, Xinyuan Song ✝ and Kai Kang ✝ (2025+). Mixed Membership Latent Variable Model with Unknown Factors, Factor Loadings and Number of Extreme Profiles. (Submitted)

6. Yue Yu, Xiangnan Feng, Kai Kang and Xinyuan Song (2025+). Bayesian nonhomogeneous Markov-switching penalized mixed-frequency regression. (Submitted)

Publications

7. Kai Kang #, Zhihao Wu #, Xinjie Qian, Xinyuan Song and Hongtu Zhu (2025). Federated Double Machine Learning for High-dimensional Semiparametric Models. 

Biometrics, accepted [pdf].

8. Kai Kang ✝, Qinxia Wang, Zhanpeng Xu and Yuanjia Wang (2025). A Bayesian Mixture of Exponential Family Factor Models for Uncovering Disease Progression Subtypes. 

Statistics in Medicine, accepted [pdf].

9. Zhihao Wu *, Xinyuan Song ✝ and Kai Kang ✝ (2025). Longitudinal Mixed Membership Image-on-Scalar Model. 

Journal of Computational and Graphical Statistics, accepted [pdf].

(An earlier version of this paper received JSM 2024 Paper Award from the ASA Biometrics Section.)

10. Yuyang He *, Xinyuan Song ✝ and Kai Kang ✝ (2024). Joint Mixed Membership Modeling of Multivariate Longitudinal and Survival Data for Learning the Individualized Disease Progression. 

The Annals of Applied Statistics, 18(3), 1924-1946 [pdf].

11. Kai Kang and Xinyuan Song (2023). Joint Modeling of Longitudinal Imaging and Survival Data. 

Journal of Computational and Graphical Statistics, 32, 402-412 [pdf].

12. Kai Kang, Deng Pan and Xinyuan Song (2022). A Joint Model for Multivariate Longitudinal and Survival Data to Discover the Conversion to Alzheimer’s Disease. 

Statistics in Medicine, 41, 356-373 [pdf]

13. Kai Kang and Xinyuan Song (2022). Consistent Estimation of Joint Model for Multivariate Longitudinal and Survival Data with Latent Variables. 

Journal of Multivariate Analysis, 187, 104827 [pdf]

14. Xiaoxiao Zhou, Kai Kang, Timothy Kwok and Xinyuan Song (2021). Joint Hidden Markov Model for longitudinal and Time-to-Event Data with Latent Variables. 

Multivariate Behavioral Research, published online [pdf]

15. Xiaoxiao Zhou, Kai Kang and Xinyuan Song (2020). Two-part Hidden Markov Models for Semicontinuous Longitudinal Data with Nonignorable Missing Covariates. 

Statistics in Medicine, 39, 1801-1816 [pdf]

16. Kai Kang, Xinyuan Song, Xiaoqiong Hu and Hongtu Zhu (2019). Bayesian Adaptive Group Lasso with Semiparametric Hidden Markov Models. 

Statistics in Medicine, 38(9), 1634-1650 [pdf]

17. Kai Kang, Jingheng Cai, Hongtu Zhu and Xinyuan Song (2019). Bayesian Hidden Markov Models for Delineating the Pathology of Alzheimer's Disease. 

Statistical Methods in Medical Research, 28(7), 2112-2124 [pdf]

18. Deng Pan, Kai Kang, Chunjie Wang and Xinyuan Song (2019). Bayesian Proportional Hazards Model with Latent Variables. 

Statistical Methods in Medical Research, 28(4), 986-1002 [pdf]

19. Jingheng Cai, Ming Ouyang, Kai Kang and Xinyuan Song (2018). Bayesian diagnostics of hidden Markov structural equation models with missing data. 

Multivariate Behavioral Research, 53(2), 151-171 [pdf]

20. Xinyuan Song, Kai Kang, Ming Ouyang, Xuejun Jiang and Jingheng Cai (2017). Bayesian Analysis of Semiparametric Hidden Markov Models with Latent Variables.

Structural Equation Modeling - A Multidisciplinary Journal, 25(1), 1-20 [pdf]