For the 19th year, Penn State’s Center for Astrostatistics is offering its week-long Summer School in statistical methodology for astronomy. The 19th summer school will be held virtually via Zoom from June 3-7, 2024. It features

  • Pre-recorded lectures & live Q&A sessions via Zoom.
  • Computational lab tutorials as homework to complement each lesson.
  • Opportunities to ask further questions to instructors, teaching assistants and other students via Slack.
  • Parallel sessions for learning various data analytic tools.

The School provides an intensive program in statistical inference covering topics such as principles of probability and inference, regression and model selection, bootstrap resampling, supervised / unsupervised learning, Bayesian data analysis, Markov chain Monte Carlo (MCMC), nested sampling, time series analysis, spatial statistics, deep learning neural networks, Gaussian process, and random forest.  

Lectures will be prepared by experienced instructors in astrostatistics and will be scheduled for watching from 10:00am-4:00pm EDT (UTC-4 hours) with live Q&A session(s) each day. Lab tutorials in the form of computational notebooks will be provided as homework to reinforce the learning experience, encouraging  participants to exercise the methods with astronomical datasets illustrating realistic challenges faced in contemporary research. We anticipate attendees will be spread across many timezones. Attendees are strongly encouraged to work through tutorials either in the afternoon/evening following the lesson or during the morning before the next day’s live lessons begins, and are encouraged to engage with faculty, teaching assistants and other students via Slack as they work on the lab tutorials. Attendees can ask questions about lectures and/or lab tutorials anytime via Slack.

Intended Audience & Preparation 

The School is designed for graduate students and postdocs in Astronomy & Astrophysics.  Advanced undergraduate students and more senior researchers are also welcome to join.  The Summer School in Statistics for Astronomers assumes working mathematical knowledge at the level of undergraduate physics or astronomy major, but does not assume that students have had any prior formal training in statistics.

The School assumes that participants have basic familiarity with programming, but does not assume that participants are familiar with the syntax of any particular language.  The School will provide training in using R, an open-source statistical software environment widely used among statisticians.  It is highly recommended that all of the participants install R before the beginning of the summer school. Click here to download R on your local machine.  Participants will also be provided instructions for installing R as a part of the first day tutorial, so they can complete notebook-based labs on their local system.

Serious participants should expect to engage for full-days during the week.  In addition to lectures and tutorials, we recommend budgeting ~2 hours per day to work through the computational notebook tutorials, and consulting with instructors, teaching assistant & other participants on Slack.  While this may be a heavier expectation than is common for many other meetings, previous years’ experience teaches us that applying lessons learned in lab tutorials is extremely valuable for participants to gain intuition for methods and improve their data analysis skills.

Course Information

Tentative schedule is available here.

List of instructors:

  • Suzanne Aigrain (Oxford University) for Gaussian process
  • Jogesh G. Babu (Penn State University) for model fitting, validation, and bootstrapping
  • David Banks (Duke University) for deep learning neural network
  • Eric Feigelson (Penn State University) for time series and R tutorials
  • Peter Freeman (Carnegie Mellon University) for random forest
  • Murali Haran (Penn State University) for Markov chain Monte Carlo and spatial statistics
  • David Hunter (Penn State University) for probability
  • Tom Loredo (Cornell University) for Bayesian inference
  • Joshua Speagle (University of Toronto) for nested sampling
  • Hyungsuk Tak (Penn State University) for statistical inference
  • Ashley Villar (Harvard University) for regression, supervised and unsupervised learning

Registration Information

Registration Deadline: May 11th, 2024

Registration Fee: $150

Refunds and Cancellations:
Refunds will be issued for cancellation by the registrant if we receive your notification in writing by May 11th, 2024.

A $50 administrative fee will be charged for all cancellations. Refund requests made after that time will not be honored.

The University may cancel or postpone any course or activity because of insufficient enrollment or other unforeseen circumstances. If a program is canceled or postponed, the University will refund the full registration fees but cannot be held responsible for any other related costs, charges, or expenses. 

Register Now

Contact

For any questions about the virtual summer school, please feel free to reach out to the Penn State Center for Astrostatistics (cast@psu.edu).