Skip to main content
Cornell university
Cornell Statistics and Data Science Cornell Statistics and Data Science
  • About Us

    About Us
    Cornell's Department of Statistics and Data Science offers four programs at the undergraduate and graduate levels. Undergraduates can earn a BA in statistical science, social...

    Welcome to the Department of Statistics and Data Science
    History
    Facilities
    Statistics Graduate Society
    Recently Published Papers
  • Academics

    Academics

    Undergraduate
    PhD
    MPS
    PhD Minor in Data Science
    Courses & Course Enrollment
  • People

    People

    Faculty
    Field Faculty
    PhDs
    Emeritus Faculty
    Academic Staff
    Staff
    Research Areas of Expertise
    Statistical Consultants
  • News and Events

    News and Events

    Events
    News
  • Resources

    Resources

    Professional Societies and Meetings
    Affiliated Groups
    Career Services
    Cornell Statistical Consulting Unit
  • Alumni

    Alumni
    Cornell's Statistics and Data Science degrees prepare students for a wide variety of careers, from academia to industry.  See the After Graduation page for a general overview of...

    Alumni Profiles

Search form

You are here

  1. Home 
  2. Events 
  3. Statistics Seminars

Statistics Seminar Speaker: Josh North 2/7/2024

Event Layout

Wednesday Feb 07 2024

Statistics Seminar Speaker: Josh North 2/7/2024

4:15pm @ G01 Biotech
In Statistics Seminars

Josh is currently a postdoctoral fellow in the CASCADE research group at Lawrence Berkeley National Laboratory. Prior to his postdoc, he completed his Ph.D. at the University of Missouri Columbia with Drs. Christopher Wikle and Erin Schliep. His main area of research is spatio-temporal statistical modeling with an emphasis on environmental and atmospheric applications.

Talk: Data-driven dynamical modeling for spatio-temporal statistics

Abstract: Classically, spatio-temporal data can be modeled using a mixed-effect representation where the fixed-effect captures the mean behavior and the random-effect describes the residual structure in the data. One approach for modeling each component statistically is to use dynamical equations parameterized with differential equations for the fixed-effect and a structured, low-rank representation for the random-effect: this approach encodes physical processes into the statistical model while enabling scalability to the large data sets that are now commonplace. While some of the dynamical relationships may be known, for real-world data sets we rarely know the complete form of the governing differential equations. Additionally, the low-rank model for the structured residual is typically based on some fixed decomposition that is specified a-priori to model estimation. Here, I present a fully Bayesian data-driven approach for discovering the form of dynamical equations parameterized by differential equations with uncertainty quantification. I illustrate the method’s ability to recover the synthetic Burgers’ equation and apply the method to infer the temporal evolution of the vorticity of the streamfunction. I will also present a data-driven approach to estimate the structure of a low-rank model representation parameterized by singular value decomposition where the basis functions from the decomposition are modeled dependently and with uncertainty quantification. I highlight the efficacy of the proposed low-rank approach on synthetic examples and apply it to sea surface temperature data over the northern Pacific.

Event Categories

  • Statistics Seminars
  • Special Events

Image Gallery

A color photo of a man smiling for a photo
  • Home
  • About Us
  • Contact Us
  • Careers
© Cornell University Department of Statistics and Data Science

1198 Comstock Hall, 129 Garden Ave., Ithaca, NY 14853

Social Menu

  • Facebook
  • Twitter
  • YouTube
Cornell Bowers CIS College of Computing and Information Science Cornell CALS ILR School

If you have a disability and are having trouble accessing information on this website or need materials in an alternate format, contact web-accessibility@cornell.edu for assistance.