Wednesdays 4:30 pm at SAMSI, Research Triangle Park, NC, beginning Wednesday, September 7, 2016
No class during week of Thanksgiving: Wednesday November 23, 2016
Last class: Wednesday, November 30, 2016
Instructors: This course will be jointly taught by few visitors at SAMSI with James Long (Texas A&M, Statistics, email@example.com) as the lead instructor.
Course Description: With the advance of digital imaging techniques, astronomy has become a data science in which knowledge creation depends on applying and developing sophisticated statistical methodology to large and/or complex data sets. This course will cover common types of data in astronomy such as light curves, spectra, and images as well as statistical methods used for analyzing these data sets, such as functional data analysis, measurement error models, hierarchical models, survival analysis, and machine learning techniques. An emphasis will be placed on the complexity of the inference tasks faced by astronomers and the propagation of uncertainty across several levels of inference. Guest lecturers will discuss topical issues in the analysis of astronomy data. Students will complete a final project in groups based on data or statistical methodology presented during the course.
The course will be aimed at a wide audience in an effort to appeal to students with either an astronomy or statistics background. While there are no formal prerequisites students will benefit from having some past experience with medium to large data sets and a familiarity with statistical methods such as maximum likelihood and regression. Theory will be kept to a minimum with an emphasis instead on astronomy data and statistical methodology.
Textbook / Reading: No textbook is required.
Some useful references:
Feigelson and Babu: Modern Statistical Methods for Astronomy, ISBN 9780521767279.
Ivezic, Connolly, VanderPlas, and Gray: Statistics, Data Mining, and Machine Learning in Astronomy, ISBN 9780691151687.
Sampling of possible articles for class discussion:
- A Framework for Statistical Inference in Astrophysics” Ann. Rev. of Stat., Schafer
- Some Aspects of Measurement Error in Linear Regression of Astronomical Data” ApJ, Kelly
- Unsupervised Transient Light Curve Analysis via Hierarchical Bayesian Inference” ApJ, Sanders
- Modeling Light Curves for Improved Classication” Statistical Analysis and Data Mining, Faraway
- Multilevel Bayesian Framework for Modeling the Production, Propagation and Detection of Ultra-
- High Energy Cosmic Rays” Annals of Applied Statistics, Soiaporn
Grading: There will be no exams. The grade is determined by the completion of four sets of homework.
Registration: (processed through the respective university)
- UNC-CH: STOR 894.002 and MATH 892.002
- Duke: STA 790.01 and MATH 790-71-01
- NCSU: ST 810-002 and MA 810-002
**To view video presentations from this course, please CLICK HERE
Questions: email firstname.lastname@example.org