URL: http://hea-www.harvard.edu/AstroStat/CAS2010

The California-Boston-Smithsonian Astrostatistics Collaboration plans to host a mini-workshop on Computational Astro-statistics. With the advent of new missions like the Solar Dynamic Observatory (SDO), Panoramic Survey and Rapid Response (Pan-STARRS) and Large Synoptic Survey (LSST), astronomical data collection is fast outpacing our capacity to analyze them. Astrostatistical effort has generally focused on principled analysis of individual observations, on one or a few sources at a time. But the new era of data intensive observational astronomy forces us to consider combining multiple datasets and infer parameters that are common to entire populations. Many astronomers really want to use every data point and even non-detections, but this becomes problematic for many statistical techniques.

The goal of the Workshop is to explore new problems in Astronomical data analysis that arise from data complexity. Our focus is on problems that have generally been considered intractable due to insufficient computational power or inefficient algorithms, but are now becoming tractable. Examples of such problems include: accounting for uncertainties in instrument calibration; classification, regression, and density estimations of massive data sets that may be truncated and contaminated with measurement errors and outliers; and designing statistical emulators to efficiently approximate the output from complex astrophysical computer models and simulations, thus making statistical inference on them tractable. We aim to present some issues to the statisticians and clarify difficulties with the currently used methodologies, e.g. MCMC methods. The Workshop will consist of review talks on current Statistical methods by Statisticians, descriptions of data analysis issues by astronomers, and open discussions between Astronomers and Statisticians. We hope to define a path for development of new algorithms that target specific issues, designed to help with applications to SDO, Pan-STARRS, LSST, and other survey data.

We hope you will be able to attend the workshop and present a brief talk on the scope of the data analysis problem that you confront in your project. The workshop will have presentations in the morning sessions, followed by a discussion session in the afternoons of both days.

Abstracts are due Sep 17.

Meeting Justification

We propose to highlight the growing use of ‘non-parametric’ techniques to distill meaningful science from today’s astronomical data. Challenges range from Kuiper objects to cosmology. We have chosen just a few ‘teaching’ examples from this lively interdisciplinary area.

Meeting Notes

This ‘Astro-Statistics’ special session is proposed in concert with an ‘Astro-Informatics’ Special Session, organized by Kirk Bourne. In this proposed ‘Non-Parametrics for the Non-Specialist’ session, we are highlighting just a few of the new, outstanding, applications. Many are coming to fruition just now, in this age of large data-sets, complex instruments, and subtleties of distilling accurate science from indirect measurements. We chose to highlight: complex models (cosmology, black hole mass distributions); and complex data, such as image (spatial); and timing analyses (e.g. transients such as the distribution of Kuiper objecs) from surveys. We invited a mixture of newer and seasoned speakers; and ones that will make good ‘teaching examples’. At the same time, we left out many new areas. Hence we are planning a lively, associated, poster session. The format will be: An Intro by one of the seasoned statisticians; followed by ‘examples’ talks by two astronomers and a physicist. Following, another of the senior statisticians will discuss the principles. Finally, a senior astrophysicist will summarize challenges for the future. We plan to leave time for one-minute poster advertisements highlighting other areas. Expected participants include: Eric Feigelson, Brandon Kelly, Meyer Pesenson, Stanislav (George) Djorgovski, Tom Loredo, Alanna Connors, Pavlos Protopapas, Katrin Heitmann, Chad Schaefer, Xiao Li Meng.

Owen talks in detail about how the Copernican model came to supplant the Ptolemaic model. In particular, he describes how Kepler went from Ptolemaic epicycles to elliptical orbits. Contrary to general impression, Kepler did not fit ellipses to Tycho Brahe’s observations of Mars. The ellipticity is far too small for it to be fittable! But rather, he used logical reasoning to first offset Earth’s epicyle away from the center in order to avoid the so-called Martian Catastrophe, and then used the phenomenological constraint of the law of equal areas to infer that the path must be an ellipse.

This process, along with Galileo’s advocacy for the heliocentric system, demonstrates a telling fact about how Astrophysics is done in practice. Hyunsook once lamented that astronomers seem to be rather trigger happy with correlations and regressions, and everyone knows they don’t constitute proof of anything, so why do they do it? Owen says about 39 1/2 minutes into the lecture:

Here we have the fourth of the myths, that Galileo’s telescopic observations finally proved the motion of the earth and thereby, at last, established the truth of the Copernican system.

What I want to assure you is that, in general, science does not operate by proofs. You hear that an awful lot, about science looking for propositions that can be falsified, that proof plays this big role.. uh-uh. It is coherence of explanation, understanding things that are well-knit together; the broader the framework of knitting the things together, the more we are able to believe it.

Exactly! We build models, often with little justification in terms of experimental proof, and muddle along trying to make it fit into a coherent narrative. This is why statistics is looked upon with suspicion among astronomers, and why for centuries our mantra has been “if it takes statistics to prove it, it isn’t real!”

The meeting website is at: http://www.confcon.com/head2008/.The abstract submission deadline is January 25, 2008 (now past, but late abstracts are not unheard of among astronomers).

If you are attending the meeting, and plan to present posters or talks that deal with astrostatistical methods or techniques, we welcome you to participate in this session. When you submit an abstract, be sure to indicate a category of “Other” and in the comments field state that it belongs with the AstroStatistics special session.If you have questions, please contact Aneta or me. There is also a page for this session on the astrostat google groups site.

Update (1/22): The abstract submission page currently says that only one abstract is allowed per person. We have been informed that this is incorrect, and that people can submit two abstracts, one for the special session and one as a regular contribution. Note that posters will be up only one day, and those associated with a special session will be put up the day of the session.

Update (1/26): A detailed program is not yet available, but here is a description of the session:

Astrostatistics: Methods and Techniques

This session will provide a forum for the discussion and presentation of statistical challenges in high energy astrophysics, highlighting the great deal of progress that has been made in methods and techniques over the past decade. The one hour session will cover the current and future directions in Astrostatistics, and will include a discussion of MCMC methods in the context of specific applications (such as propagating calibration errors, defining the significance of image features, etc.); a discussion of standardized methods for computing detection limits, upper limits, and confidence intervals for weak sources; and hypothesis testing and its limitations (including the significance testing of emission lines).

Update (2/19): We have been allocated the mid-day slot of March 31. The session will run from 12:30pm till 1:30pm2pm. The tentative program is as follows:

• Remarks on current and future trends in AstroStatistics, by Eric Feigelson
• Poster haiku
• F-Test theory and usage, by David van Dyk
• Discussion on MCMC techniques, led by Andy Ptak

Update (2/26): The final program is out, and the AstroStat session is scheduled for 12:30pm-2pm at the Museum/Bunker Hill Room.

Update (4/1): The talks and posters associated with the AstroStat special session are now online at
http://hea-www.harvard.edu/AstroStat/HEAD2008/. Additional comments and descriptions will be archived there.

• [astro-ph:0710.2371]
  Probing Non-Gaussianity In The Cosmic Microwave Background Anisotropies: One Point Distribution Function by E. Jeong, and G. F. Smoot
• [astro-ph:0710.2397]
  Initial-Final Mass Relationship for Stars of Different Metallicities by X. Meng and Z. Chen
• [astro-ph:0710.2436]
  Optical afterglows of gamma-ray bursts: a bimodal distribution?” N. Marco, G. Gabriele, and G. Giancarlo.
• [astro-ph:0710.2455]
  Superfluid turbulence and pulsar glitch statistics by A. Melatos and C. Peralta
• [astro-ph:0710.2664]
  Reconstructing Images from Projections Using the Maximum-Entropy Method. Numerical Simulations of Low-Aspect Astrotomography by A. T. Bajkova
• [astro-ph:0710.2783]
  Void Statistics and Void Galaxies in the 2dFGRS by A. M. von Benda-Beckmann and V. Mueller
• [astro-ph:0710.2872] summer school material
  Geometrical constraints on dark energy models by R. Lazkoz
• [astro-ph:0710.2882]
  The IMF of the massive star-forming region NGC 3603 from NIR AO observations by Y. Harayama, F. Eisenhauer, and F. Martins
• [hep-ph:0710.1952]
  Global neutrino parameter estimation using Markov Chain Monte Carlo by S. Hannestad
• [astro-ph:0710.3099]*
  Short Gamma Ray Bursts: a bimodal origin? by R. Salvaterra, A. Cerutti, G. Chincarini, M. Colpi, C. Guidorzi, P.Romano
• [astro-ph:0710.2397]
  Clues to Globular Cluster Evolution from Multiwavelength Observations of Extragalactic Systems by A. Kundu, T. J. Maccarone, and S. E. Zepf
  (I’ve been keen on globular clusters, best archeological objects in the universe to understand galaxy formation and the history of the universe. They are like genomes of a galaxy.)
• [stat.TH:0710.3478]
  Nonparametric estimation of a point-spread function in multivariate problems by P. Hall and P. Qiu

[1]: Personally, point spread function (PSF) plays a major role in extracting astrophysical information from raw observations.

[2]: In addition to this week’s GRB paper, [astro-ph:0710.3099]*, there were some statistical studies about multi-modality in GRBs [see Three classes of GRBs].

I think that (astro)physics provides templates such as likelihoods for a statistical analysis. The statistical disputes between the bimodality and the trimodality of GRB distribution can be settled from 1. more studies like [astro-ph:0710.3099]* or 2. better statistical inference tools or model selections. Unfortunately, 2 requires templates, which are acquired from in-depth cross matching survey studies.