Qingkai Dong 董庆凯
Hello world!!
I’m Qingkai(pronounced as Tsing-kai) Dong. This is my personal academic
website. Currently I’m a PhD student in
Statistics in UConn. My research interests
include massive data subsampling, frequentist model averaging and survival data
analysis. I’m interested in programming, too.
Education
University of Connecticut
- Ph.D. in Statistics
- Advisor: Prof. HaiYing Wang and Prof. Jun Yan
- Selected Coursework: Linear Model, Applied Statistics, Mathematical Statistics.
- Location: Storrs, CT, USA
- Duration: Sep 2023 - present
Zhongnan University of Economics and Law
- Master of Science in Mathematical Statistics
- Selected Coursework: Advanced Probability Theory, Advanced Mathematical Statistics, Survival Analysis, Non-parametric Statistics, High-dimensional Statistics
- Thesis: “Variable Selection and Model Averaging Methods of Accelerated Failure Time Models”
- Advisor: Prof. Hui Zhao
- Location: Wuhan, China
- Duration: Sep 2020 - Jun 2023
Qingdao University
- Bachelor of Science in Applied Statistics
- Selected Coursework: Real Analysis, Advanced Algebra, Probability Theory, Mathematical Statistics, Numerical Optimization, Regression Analysis, Matlab, C++
- Thesis: “Automatic Marking System based on Convolutional Neural Networks”
- Advisor: Dr. Yang Yu
- Location: Qingdao, China
- Duration: Sep 2016 - Jun 2020
Software
Balancing computational and statistical efficiency, subsampling techniques offer
a practical solution for handling large-scale data analysis. Subsampling methods
enhance statistical modeling for massive datasets by efficiently drawing
representative subsamples from full dataset based on tailored sampling
probabilities. These probabilities are optimized for specific goals, such as
minimizing the variance of coefficient estimates or reducing prediction error.
Publications
[1]
Dong Q., Liu B., Zhao H. (2023).
Weighted Least Squares Model Averaging for Accelerated Failure Time Models.
Computational Statistics and Data Analysis, 184: 107743. PDF
[2] zha
Zhao H., Liu B., Dong Q., Zhang X. (2023).
The Jackknife Model Averaging of Accelerated Failure Time Model with Current Status Data.
Acta Mathematicae Applicatae Sinica, 46(3): 313-328. (in Chinese) PDF
[3]
Zhao H., Dong Q. (2022).
A Variable Selection Method for the Additive Hazards Model with Current Status Data.
Journal of Systems Science and Mathematical Sciences, 42(5): 1314-1329. (in Chinese) PDF
Experience
Research Assistant, University of Connecticut and Servier
- Duration: Mar 2024 - present
- Description: A project on the application of machine learning methods in
adaptive clinical trial design. It includes the development of models for
predicting survival times of censored individuals and their application to
improve the accuracy of conditional power estimates for interim analyses.
Research Assistant, University of Connecticut
- Duration: Sep 2023 - present
- Description: Summarized literature on subsampling methods in statistical
models. Developed an R package which provides optimal subsampling methods for
various statistical models such as generalized linear models, softmax
regression, rare event logistic regression and quantile regression model.
Research Assistant, University of Connecticut
- Duration: Jul 2024 - Aug 2024
- Description: A project about the impact of social determinants of health
(SDOH) and frailty index on accelerated aging in breast cancer patients. It
includes pre-processing the data and performing statistical analyses on the All
of Us platform.
Teaching Assistant, Zhongnan University of Economics and Law
- Course Title: Probability Theory
- Duration: Sep 2020 - Jan 2021
- Description: Assisted in the undergraduate course Probability Theory for
60+ students, graded homework and exams.
Awards
- First-class Scholarship, Zhongnan University of Economics and Law, Sep 2022 & Sep 2020
- Second-class Scholarship, Qingdao University, Sep 2019
- First Prize, CUMCM (Contemporary Undergraduate Mathematical Contest in Modeling), Oct 2018
- First Prize, CSEE Cup (Electrical Math Modeling Competition), May 2018
Skills
- Programming Languages: R, Python, Matlab, C++, Latex
- Languages: English (fluent), Chinese (native)