Last Updated: 2008jan22


Topics in Astrostatistics

Statistics 310, Fall/Winter 2007-2008

Harvard University

Instructor Prof. Meng Xiao Li
Schedule Tuesdays 11:30 AM
Location Science Center Rm 706

Alanna Connors (Eureka Sci), Pavlos Protopapas (IIC/CfA)
18 Sep 2007
SciCen 111 706
Interdisciplinary Astronomy, Physics, and Statistics
The California-Harvard AstroStatistics Collaboration is now in its tenth year (*). We will set the framework by briefly introducing the projects from the past, and showing examples of current challenges. The statistics challenges range from deceptively simple (confidence limits on ratios and rates of very low count Poisson processes) to many-layered (understanding multi-scale images, spectra, time-variations from complicated instrumental measurements of the sky). The applications can be literally cosmic (structure and evolution of matter and the universe); and the pictures spectacular. Come join our free-wheeling discussions of the growing boundaries of twenty-first century astronomy, physics, and statistics.
(*) In our understanding, we are currently the longest-running interdisciplinary group in the Harvard College of Arts and Sciences. If anyone knows differently, we would be very interested in finding out!
Alanna Connors [.pdf]
25-26 Sep 2007
16 Oct 2007
23 Oct 2007
30 Oct 2007
Jack Steiner (CfA)
13 Nov 2007
27 Nov 2007
Spinning Stellar Black Holes
Astronomers have been studying the properties of black hole binaries for decades, but it is only within the last few years that it has become possible to measure black hole spin. Spin is crucially linked to many exciting frontiers of astrophysics including gravitational wave astronomy, quasar and stellar mass black hole jets, the evolution of super-massive black holes, and possibly gamma-ray bursts. Our group has been pioneering the use of X-ray continuum fitting to measure black hole spins, and has published results on four black hole X-ray binary systems. We are now in the process of developing a robust methodology for our work. I will be presenting an overview of our current methods, illustrating several statistical challenges we face, and posing a few questions to my statistically savvy colleagues.
18-19 Dec 2007
Paul Baines, et alia
05 Feb 2008
Ages of stellar populations from color-magnitude diagrams;
discussion of papers from astro-ph
26 Feb 2008
Andreas Zezas (CfA)
11 Mar 2008
11am EDT
Derivation and measurement of LogN-LogS distributions
Distributions of the number of observed sources as a function of their intensity (LogN-LogS) is one of the standard tools for studies of source populations and setting constrains on their cosmological evolution. I will briefly present a few examples to demonstrate their importance for astrophysical studies. I will discuss in detail how they are derived and the sources of bias and uncertainty in determining their shape. Finally I will present the most commonly used methods for their study and I will introduce the concept of a Bayesian fitting method developed by the Astrostatistics group (mainly Nondas Sourlas and David van Dyk).
Presentation [ppt]
08 Apr 2008
Abstract: To discuss some pressing astrostatistical problems that may be of interest to stat grad students.
Tom Aldcroft
22 Apr 2008
X-ray Stacking
Abstract: X-ray stacking provides a way to dig deeper into the ever-growing archives of X-ray data and estimate the mean properties of sources that are too faint to detect individually. My interest in this technique comes from the Chandra Multiwavelength Project (ChaMP), which is a very large archival survey that covers over 30 sq. degrees with good optical coverage (dedicated optical imaging / spectroscopy and Sloan Digital Sky Survey). Unfortunately the heterogeneity of the ChaMP complicates the process so I'm exploring some variations on the stacking theme. My talk will start with an overview of the technique as commonly practiced by astronomers and then discuss potential problems and a Monte-Carlo method for estimating the derived parameter distributions. Then I'll talk about stacking in the ChaMP and bring up some simple ideas for estimating physically interesting parameters using this dataset.
06 May 2008
stacking, error propagation on non-linear integrals, etc.
Fall/Winter 2004-2005
Siemiginowska, A. / Connors, A. / Kashyap, V. / Zezas, A. / Devor, J. / Drake, J. / Kolaczyk, E. / Izem, R. / Kang, H. / Yu, Y. / van Dyk, D.
Fall/Winter 2005-2006
van Dyk, D. / Ratner, M. / Jin, J. / Park, T. / CCW / Zezas, A. / Hong, J. / Siemiginowska, A. & Kashyap, V. / Meng, X.-L.
Fall/Winter 2006-2007
Lee, H. / Connors, A. / Protopapas, P. / McDowell, J., / Izem, R. / Blondin, S. / Lee, H. / Zezas, A., & Lee, H. / Liu, J.C. / van Dyk, D. / Rice, J.
Fall/Winter 2007-2008
Connors, A., & Protopapas, P. / Steiner, J. / Baines, P. / Zezas, A. / Aldcroft, T.