I am Shijie Wang, currenlty a fifth-year Ph.D. student in the Department of Statistics at University of South Carolina under supervision of Prof. Ray Bai. My primary research focuses on utilizing deep learning to address, improve and facitilite statistical inference in the following classcial statistics models:

  • Bootstrap uncertainty quantification
  • Nonparametric conditional density estimation and nonparametric multi-quantile estimation
  • Emprical Bayes and nonparameteric maximum likelihood estimation
  • Bootstrap spatial-temopral process and time-varying spatial-temopral regression
  • Skewed prediction for Electronic health records (EHR)


Ph.D. in Statistics, University of South Carolina, South Carolina, 2019.08 - Present.

  • Dissertation title: New Deep Learning Approaches to Classical Statistical Problems
  • Winner of James D. Lynch Graduate Student Research Award

B.S. in Applied Statistics, Shanghai University of International Business and Economics, Shanghai, 2015.09 - 2019.06.


  • Wang, S., Shin, M., and Bai, R. (2024). Generative quantile regression with variability penalty. [Paper][Code]

    Journal of Computational and Graphical Statistics.

  • Wang, S., Shin, M., and Bai, R. (2024). Fast boostrapping nonparametric maximum likelihood for latent mixture models. [R package]

    IEEE Signal Processing Letters.

  • Shin, M., Wang, S. and Liu, J. (2023). Generative Multi-purpose Sampler for Weighted M-estimation. [Paper][R package]

    Journal of Computational and Graphical Statistics.


  • Wang, S., Chakraborty, S., Qin, Q., and Bai, R. (2024+). A comprehensive deep generative framework for mixing density estimation. [R package]

Professional Experience

  • Bayer AG

    ML Data Scientist Intern, Saint Louis, 2023.05 - 2023.08.

  • University of South Carolina

    Deep Learning Research Assistant, Columbia, 2022.05 - now.

  • Movefin Tech

    Machine Learning Data Scientist Intern, Shanghai, 2019.03 - 2019.06.