FSU Development of Meta-Analysis and Statistical Methodology

Tuesday 06/01/2021

Dr. Lifeng Lin is an assistant professor of statistics here at FSU. Lin and his students are working on meta-analysis, which combines new and existing research findings to accurately estimate the effects of certain topics of interest. One of these topics is COVID-19 and vaccine safety. Streamlining these findings “provides a very helpful tool to make it possible to discover the true effect of certain treatments” states Dr. Lin.

Developing statistical methodology is a big task the FSU Statistics Department tackles. Researchers within the department deal with large-scale data from simulations as well as the real world. Seeing how these new methods measure up against existing methods pushes Dr. Lin and his team towards new and exciting data that benefits not only the FSU research community, but also the world.

In order to rigorously test the performance of the proposed new methods, simulation studies frequently need to be replicated over 10,000 times. Simulations that would take a regular computer several months to process, while the Research Computing Center (RCC) High Performance Computing cluster can run simultaneously in one day. Working with algorithms such as the Markov chain Monte Carlo (MCMC) requires data sets to run efficiently and quickly in order to publish crucial information. The shorter computational time makes a huge impact when working with methodology.

The simulation is essential from my perspective; that’s how the RCC enables my research,” states Dr. Lin. Without the RCC, many time-consuming jobs cannot be done.