ASTRO: Time Series Methods for Astronomy


January 11 – April 26, 2017 / Wednesdays at 4:30PM – 7:00PM

Instructor: This course will be jointly taught by visitors at SAMSI with Eric Feigelson (Department of Astronomy and Astrophysics, Penn State University, as the lead instructor.

Course Description:
The course starts with an overview of variable cosmic phenomena and characteristics of astronomical time series. Classical time series analysis in the time and frequency domain for evenly spaced data will be reviewed. This includes Gaussian and Poisson processes, smoothing and interpolation, autocorrelation and autoregressive modeling, Fourier analysis, and wavelet analysis. The class then proceeds to treatments of unevenly spaced time series commonly found in astronomical datasets, again in both the time and frequency domain. Guest lectures by expert SAMSI scholars developing advanced techniques for unevenly spaced data will be featured. Throughout the course, methods will be exercised using the public domain R statistical software environment using contemporary astronomical datasets. Students will complete R-based homeworks and a personal project in time series analysis involving a dataset of their choice.

Grading: There will be no exams. The grade is determined by the completion of homework and projects.

Registration: (processed through the respective university)

  • UNC-CH: STOR 891.001/MATH 892.001
  • Duke: 790-02
  • NCSU: MA 810.001, ST 810.006

Questions: email