CADENCE Lab
Contextual Analytics of Digital & Environmental data for Neurobehavioral, Circadian & Emotional health
Statistics for the rhythms of health across space and time.
We build theory and methods for multivariate stochastic processes to learn from wearables ⌚, smartphones 📱 and the environment 📍 — and use them to understand how mental health, sleep, physical activity and circadian rhythms unfold. Led by Debangan Dey in the Department of Statistics, Texas A&M University.
What we work on
Theory & Methods
Multivariate stochastic processes in space and time, semiparametric Gaussian copulas for mixed-type data, multivariate functional PCA, dynamic structural equation models, and graphical models. Learn more →
Digital Health Technologies
Objective streams from wearables and sensors — activity, heart rate, EEG, glucose — combined with real-time self-reports of mood, energy, sleep and stress, to build digital phenotypes of mental health.
Environment & Context
Linking health dynamics to temperature, light, greenspace and the built environment through location tracking and spatial statistics, from mood disorders to youth physical activity.
- 🚨 Looking for motivated students to work on statistical and machine learning methods for analyzing data from wearables ⌚️, smartphones 📱, with contextual location information📍!
- 📄 September 2026: New preprint — Bigraphical Matérn–Whittle (BMW) Processes for Fast Inference of Big Multivariate Spatial Data on General Domains with A. Manna and C. J. Geoga.
- ⚽ 2026: New preprint — Do In-Match Hydration Breaks Alter Match Momentum? A Within-Match Case-Crossover Analysis of the 2026 FIFA World Cup. arXiv · GitHub.
- 🏆 2026: Our team has won the NIH Director’s Challenge Innovation Award for Biologic Rhythms and Environmental Contexts in Human Health: Integrating Epidemiology and Circadian Science (PI: Kathleen Merikangas, NIMH; in collaboration with NIDDK, NHLBI, NIAAA, NIA, NHGRI, Northwestern, Nathan Kline Institute, Johns Hopkins, U Colorado, Texas A&M, and UT Southwestern). Debangan Dey (Texas A&M, Consultant) will oversee development of data management and analytic strategies for multilevel data, integrating intensive individual-level data with macro environmental factors.
- ⚽ 2026: Built a FIFA World Cup 2026 Bracket Predictor with Claude — Monte Carlo bracket odds, live ticket & hotel prices, and a trip planner that ranks your 3 cheapest venues (with day-trip detection and Google Flights links) so you can finally settle the “go or not go” debate.
- 💻 2026: Released M²FPCA — an R package for Multivariate Functional Principal Component Analysis of mixed-type functional data (continuous, truncated, ordinal, and binary) via a latent Gaussian copula. Companion to SGCTools.
- 📰 2025: Published Associations between daily outdoor temperature and subjective real-time ratings of emotional states and sleep in mood disorder subtypes in Journal of Affective Disorders — Featured by Texas A&M: Warmer Days, Better Moods? It’s Complicated.
- 📄 2026: New preprint — Doubly-Unlinked Regression for Dependent Data with A. Burman and S. Choudhury.
Assistant Professor of Statistics · Principal Investigator
About the PI
Debangan Dey is an Assistant Professor in the Department of Statistics at Texas A&M University. His research focuses on building theory and methods for multivariate stochastic processes as a unifying AI framework to analyze intensive, multilevel, multimodal, longitudinal data collected across space and time. This type of data arises in studies employing Digital Health Technologies, such as smartphone apps and smartwatches, with contextual spatial information like weather, light, and greenspace etc. through location tracking. His work aims to uncover how mental health, sleep, physical activity, and the environment interact and evolve over time.
- Multivariate stochastic processes across space and time
- Mixed-type functional data and semiparametric Gaussian copulas
- Digital health technologies (wearables, EMA) and mental health