StanCon serves as the gathering point for the global Stan community—a diverse group of statisticians, data scientists, researchers, and practitioners who use Stan for Bayesian inference across a remarkable range of application domains. The conference provides a venue for sharing practical experiences, learning about new features and best practices, and strengthening the connections within the Stan ecosystem.
History
The first StanCon is held at Columbia University in New York, establishing a dedicated conference for the Stan community.
StanCon is held at Aalto University in Helsinki, Finland, and at the Asilomar Conference Grounds in California, broadening the conference's international reach.
StanCon continues to grow, with presentations spanning pharmacometrics, ecology, political science, and engineering.
StanCon returns with both in-person and virtual components, reflecting the post-pandemic evolution of academic conferences.
Conference Format
StanCon features a mix of keynote presentations by Stan core developers and prominent users, contributed talks showcasing applications and methodological innovations, tutorials and workshops on Stan features and modeling techniques, and developer sessions where the technical direction of Stan is discussed. A distinctive feature is the emphasis on case studies—detailed narratives of real-world problems solved using Stan, including the challenges encountered and lessons learned.
Stan case studies are one of the most valuable outputs of the StanCon community. These detailed, reproducible analyses demonstrate how to approach real problems using Bayesian methods in Stan, covering everything from model specification and prior choice to diagnostics and interpretation. They serve as both educational resources and templates for practitioners tackling similar problems.
Community and Culture
StanCon reflects the Stan community's distinctive culture: technically rigorous, practically oriented, and welcoming to newcomers. The conference attracts a remarkably diverse audience, from graduate students learning Bayesian methods for the first time to experienced researchers pushing the boundaries of what Stan can do. This mix creates a dynamic environment where fundamental concepts and cutting-edge developments are discussed side by side.
The conference also serves as a meeting point for the distributed Stan development team, providing an opportunity for core developers, package maintainers, and community contributors to coordinate their efforts and plan the future direction of the project.
"StanCon is not just a software conference—it is a conference about how to do Bayesian statistics well. The talks are as much about modeling and workflow as they are about code."— Andrew Gelman
Impact
StanCon has played an important role in building and sustaining the Stan community, which now numbers thousands of active users worldwide. By providing a forum for sharing experiences, identifying common challenges, and celebrating successes, the conference has helped Stan evolve from a research tool into a mature platform that supports Bayesian inference across virtually every scientific and professional domain.