Meet Research Fellow Sherilyn Long
Sherilyn is a Research Fellow in the Department of Data Science and AI at Monash University, where she also completed her PhD in Data Science.
Her research focuses on time series forecasting, Bayesian inference and machine learning, with a particular interest in developing new statistical and machine learning methods for complex forecasting problems. Her work has been applied to areas including large-scale retail forecasting and cloud resource forecasting.
Within the Temporal Analytics Lab, Sherilyn is focused on time series forecasting, Bayesian inference and machine learning. She is interested in developing forecasting tools that are not only accurate, but useful in addressing real-world problems.
One project Sherilyn is particularly proud of is her research into time series foundation models. The work formalises an important limitation of existing general-purpose time series foundation models: strong performance on broad benchmarks does not necessarily translate to strong performance in specialised domains. Motivated by this theory, the project develops a specialist foundation model for yearly forecasting, a relatively underrepresented area where datasets often contain only short historical contexts.
Sherilyn’s contribution includes developing a theoretical framework using Bayesian decision theory to understand the training objectives and limitations of foundation models. She has also helped shape the research direction and led the writing of the manuscript, which is currently in preparation. The work offers a new perspective on how time series foundation models could be designed and evaluated for specialised forecasting domains.
In 2026, Sherilyn was also selected as a recipient of the Women’s Research Accelerator Program, recognising her research and supporting her continued development as a researcher.
Looking ahead, Sherilyn sees the Temporal Analytics Lab as an opportunity to bring together researchers with complementary expertise to tackle important problems, while exploring new research areas and developing interdisciplinary collaborations.