Student Spotlight: Andrew Chin

Andrew Chin is a third-year PhD candidate in the Department of Biostatistics. Andrew previously worked as a software engineer and product manager for a healthcare data platform company, and during COVID was part of Carnegie Mellon University's COVID19 response effort in their Machine Learning Department.

• Hometown: Fremont, CA
• Previous Degrees Earned: BS in Statistics, University of California, Davis; MS in Statistics, University of California, Davis
• Current Program: Doctor of Philosophy (PhD)
• Program Entry Year: 2021
• Area of Focus: Computationally efficient Markov chain Monte Carlo (MCMC) methods for large scale data.

What led you to Hopkins and choosing to study biostatistics? 

My interest in biostatistics began with my first job at a health technology startup after completing my master’s degree, where I worked on deidentifying and linking disparate healthcare data. I saw firsthand how messy yet valuable the data from EHRs, claims, and other sources was. During COVID, I volunteered at a lab at CMU that was forecasting COVID cases and later joined full-time, which brought me closer to more health-related data and academia; I decided to pursue a PhD soon after and ended up at Hopkins.

Have you had any internships or jobs that have been helpful in your biostatistics learning journey?

My jobs before coming to Hopkins were useful in building computational and soft skills. Most of my work was software engineering, and I had the privilege to build and ship a number of products while learning from more senior programmers. My research is computation heavy, so having these skills has been extremely helpful. I also gained experience in collaboration and communication across teams. During my time here, I did an internship at Sandia National Laboratories, in their Computational Data Science Group, where I was able to add new methods to my toolkit and learn about new application areas.

What do you like most about the Biostatistics Department?

The people that make up the Department. Everyone I’ve met has been kind and generous with their time and knowledge, and the variety of backgrounds and research interests makes it a great place to broaden your knowledge. The atmosphere in the Biostatistics Department is always welcoming, and there is a strong emphasis on collaboration, both within our Department and with other departments.

The atmosphere in the Biostatistics Department is always welcoming, and there is a strong emphasis on collaboration, both within our Department and with other departments.

What has been your favorite class so far at Hopkins?

I enjoyed the Advanced Methods in Biostatistics Series my first year here. I felt like it pushed me to fill a lot of gaps in my knowledge and reinforce areas that I was weak in coming out of my master’s degree. The course covered methods which have numerous practical applications while also maintaining a good level of theoretical rigor in justifying their use.
    
Tell us about a project you are currently working on that you are excited about.

One of the methods I’m working on enables the use of existing state-of-the-art MCMC algorithms on discontinuous distributions which they traditionally could not be applied to. These distributions often arise in Bayesian variable selection and are crucial for analyzing observational health data, where only a few relevant features may exist among thousands of potential variables. Our approach enables us to fit models that would otherwise be prohibitively expensive to use on these large datasets.

Are you involved in any Working Groups? Can you share about one of them and what you enjoy about it.

I’ve been part of the Bayesian Learning and Spatio-Temporal modeling (BLAST) working group since I joined Hopkins, which is led by Abhi Datta and my advisor Aki Nishimura. We have regular seminars from both internal and external speakers, and in addition to getting to hear about different research areas, it’s been a good venue for me to gain experience presenting and get feedback in a low-pressure environment.

MEET OUR STUDENTS