STAtistical Methods for the Physical Sciences
Research Group

Many problems in the physical sciences share common statistical challenges including heterogeneous data from multiple probes, uncertainty quantification, ill-posed inverse problems, spatio-temporal data and complex simulations.

In 2018, a group of us at CMU started the STAMPS focus group on the realization that the scientific value of modern data in the physical sciences can only be fully realized by developing new statistical methods tailored to these unique challenges. We meet weekly and regularly have domain scientists give talks at our local research meetings.

Starting in the summer of 2020, STAMPS is offering public colloquia-style webinars, in addition to our local group meetings.

Applications:

Common STAMPS themes:

  • Massive, high-dimensional data sets
  • Heterogeneous data from multiple sources
  • Uncertainty quantification
  • Spatio-temporal statistics
  • Ill-posed inverse problems
  • Non-Gaussian data
  • Large-scale simulations
  • Low signal-to-noise ratio problems


  • Interested in joining the public webinar mailing list? Subscribe here!

  • Are you a student or faculty at CMU or UPitt and want to join our local STAMPS meetings? Subscribe to the group meeting mailing list!

  • Interested in giving an informal talk or brainstorm ideas and potential research projects at our local meetings? Contact Mikael Kuusela and Ann Lee.


CMU Department of Statistics & Data Science Members

Co-Directors: Mikael Kuusela and Ann Lee

Coordinator and Webmaster: Nic Dalmasso