Time Series Connect
Interdisciplinary online seminar on time series research and practice
About Us
Welcome! We are an online interdisciplinary seminar series that brings together researchers, practitioners, and students from fields such as statistics, computer science, economics, business, engineering, and other data-rich disciplines to share insights and advances in time series analysis. Everyone is welcome to join from anywhere in the world.
Format
We host a mix of academic talks, panels, and interactive sessions that connect ideas from research, industry, and government. Sessions run about 55 minutes, typically a 45-minute presentation followed by 10 minutes of discussion. We welcome multiple speakers on a shared theme as well as panel formats. In addition to academic talks, we also invite presentations from industry and public-sector professionals to showcase real-world applications of time series analysis and encourage cross-sector dialogue.
Each event often ends with a 30-minute Virtual Coffee, where participants can chat, ask questions, and connect. It might be open networking or a light interview-style chat with the speaker(s). Everyone’s welcome, even if they missed the main talk!
Getting Involved
No matter who you are or where you’re from, you’re welcome to sign up to save upcoming events to your calendar and connect with us via the Google form or at timeseriesconnect@gmail.com.
UConn students can contribute educational blog posts and assist with organizing seminar activities: learn more here.
Bio, Abstract & Materials
About the speaker:
Peter F. Craigmile is a Professor in the Department of Mathematics and Statistics at Hunter College, CUNY. His research interests include time series and longitudinal analysis, spatial statistics, and spatio-temporal modeling. He is currently interested in developing statistical methodologies for the statistical inference and theory surrounding stationary and nonstationary non-Gaussian spatial and temporal processes. He works on building scientifically relevant hierarchical statistical models, applied to areas such as climate science, environmental health, psychology, and public health.
Abstract:
Methods of estimation and forecasting for stationary models are well known in classical time series analysis. However, stationarity is an idealization which, in practice, can at best hold as an approximation, but for many time series may be an unrealistic assumption. We define a class of locally stationary processes which can lead to more accurate uncertainty quantification over making an invalid assumption of stationarity. This class of processes assumes the model parameters to be time-varying and parameterizes them in terms of a transformation of basis functions that ensures that the processes are locally stationary. We develop methods and theory for parameter estimation in this class of models, and propose a test that allow us to examine certain departures from stationarity. We assess our methods using simulation studies and apply these techniques to the analysis of an electroencephalogram time series. This is joint research with Shreyan Ganguly.
Bio, Abstract & Materials
About the speaker:
Hie Joo Ahn is a Principal Economist at the Federal Reserve Board’s Division of Research and Statistics, where she works in the Current Macroeconomic Conditions section. She joined the Board in 2015 after receiving her Ph.D. in Economics from the University of California, San Diego, and previously received a B.A. in Business Administration from Seoul National University. Before her doctoral studies, she worked at the Bank of Korea (the Korean central bank) conducting policy work on inflation forecasting, monetary policy, and international capital flows. Her research focuses on macroeconomics, labor market dynamics, time-series econometrics, forecasting, and nowcasting, and she has published on topics such as unemployment dynamics, inflation, and monetary policy. Hie Joo also serves as a member of the Conference on Research in Income and Wealth at the NBER.
Abstract:
TBA
Bio, Abstract & Materials
About the speaker:
Jingchao Ni is an Assistant Professor in the Department of Computer Science at the University of Houston. Prior to this, he was a researcher at the Data Science Department of NEC Labs from 2018 to 2022 and the AWS AI Labs from 2022 to 2024. He received his Ph.D. degree from College of IST, The Pennsylvania State University in 2018, advised by Prof. Xiang Zhang. His research is centered around machine learning, data mining, and artificial intelligence, with a focus on time series analysis through cross-modal learning, multimodal integration, and LLM reasoning. His research has been extended to applications in healthcare (including personalized healthcare, press coverage: Science Japan, KeguanJP), biomedicine, cyber-physical systems, and AIOps (e.g., deployed in AWS cloud systems), and published in refereed conferences (e.g., ICLR, ICML, NeurIPS, ACL, AAAI, CVPR, KDD, WWW) and journals (e.g., IEEE TKDE, ACM TKDD), with more than 20 patents filed or granted.
Abstract:
TBA
