Skip to content. | Skip to navigation

Personal tools

Navigation

You are here: Home / ASAIP Editors' Blog / ASAIP Editors' Blog: April 2015

ASAIP Editors' Blog: April 2015

This month's blog features statements from several leaders of organizations involved in astrostatistics and astroinformatics.

Astrostatistics and Astroinformatics Portal

Greetings from the ASAIP Editors,

In this blog, we are very pleased to invite brief statements from several leaders of organizations or media that provide resources or forums for researchers -- astronomers, statisticians and information scientists -- involved in astrostatistics and astroinformatics.  Alicem, Emille, Rafael, Erzsebet and Jessi are all donating their energy to different facets of the broader enterprise.

But first, as the summer time approaches, we highlight a few upcoming workshops and conferences that may interest you.  A full listing can be found here

  • Hot-wiring the Transient Universe IV (12-15 May, Santa Barbara CA USA)
  • SAMSI/SAVI Indo-US Workshop on Time Series Analysis (25-30 May, Pune IN)
  • The Local Group astrostatistics conference (1-4 June, Ann Arbor MI USA)
  • SciCoder 6 Workshop (8-12 June, New York NY USA)
  • Methodological advances in statistics related to BIG DATA (8-12 June, Castro Urdiales ES)
  • useR! Conference (30 June - 3 July, Aalborg DK)
  • Astrostatistics at the 60th World Statistics Congress (26-31 July, Rio de Janeiro BR)
  • Mocking the Universe: Better Science through Data Simulation (27-29 July, Baltimore MD USA)
  • Statistics and Exoplanets (3-6 August, Honolulu HI)
  • Joint Statistical meeting JSM 2015 (8-13 August, Seattle WA USA)
  • Third La Serena School for Data Science: Applied Tools for Astronomy (15-22 August, La Serena CL)

 

Yours, 

Eric Feigelson & Joe Hilbe 

++++++++++++++++++++++++++++++

Alice Allen is editor of the Astrophysics Software Code Library and Chair of the new American Astronomical Society Special Interest Group on software publication.

The Astrophysics Source Code Library (ASCL) is a free online registry for source codes used in research in, or submitted to, peer-reviewed publications. The ASCL currently has over 1000 code entries and is indexed by the SAO/NASA Astrophysics Data System (ADS). Its entries are citable by using the unique ascl ID assigned to each code, and each code entry's can be found by prefacing the number with ascl.net (i.e., ascl.net/1201.001).

A focus of the volunteer resource's goals for 2015 is working with the community to improve software citation in journals. Elsevier's Astronomy & Computing has been identifying codes by their ASCL IDs almost since its inception. Springer's new journal Computational Astrophysics & Cosmology has also expressed interest in using ASCL IDs, and this spring, the ASCL is participating in a workshop at GitHub, funded by the Alfred P. Sloan Foundation, with ADS, the AAS and its publisher IoP, and others on software discovery issues. Earlier this year, the ASCL arranged a Special Interest Group meeting on software publication at the AAS meeting in Seattle which included publishers, editors, and researchers which we expect to result in recommendations for authors, journals, the ASCL, and researchers to better enable code citation and discovery. 

+++++++++++++++++++++++++++++

Emille Ishida is moderator of a new Facebook page that highlights material in the Astrostatistics and Astroinformatics Portal.  

Last December the Astrostatistics and Astroinformatics Portal released its Facebook counterpart at www.facebook.com/asaip.psu.edu.  The goal is to serve as another mechanism to advertise the capabilities of the portal and the activities of the International Astrostatistics Association.  Daily updates are posted  covering the IAA, activities of its members,  and funny facts about statistics and astronomy. The page has 350+ followers.  If you are on Facebook, you are welcome to join us! 

+++++++++++++++++++++++++++++

Rafael de Souza is organizer of the Cosmostatistics Initiative (COIN) tool development group within the International Astrostatisics Association.

The COsmostatistics INitiative (COIN), a working group built within the International Astrostatistics Association (IAA), aims to create a friendly environment where hands-on collaboration between astronomers, cosmologists, statisticians and machine learning experts can flourish. COIN is designed to promote the development of a new family of tools for data exploration in cosmology.

One recent accomplishment is a series of studies aimed to spread the use of generalized linear models within the astronomical community.  Two papers are now accepted in the Astronomy and Computing journal: The Overlooked Potential of Generalized Linear Models in Astronomy - I: Binomial Regression and II: Gamma regression and photometric redshifts

COIN also seeks to facilitate the use of contemporary exploratory and visualization techniques commonly used in other scientific fields but not fully exploited in astronomy through a GitHub software toolbox at cointoolbox.github.io.  One package recently developed is AMADA package, Analysis of Multidimensional Astronomical DAtasets. The code, using the elegant Shiny interface to the R statistical software environment,  allows the user to visualize subgroups of variables with high association in a hierarchical tree structure through diverse visual tools, such as graphs, chord diagrams, dendrograms and heatmaps. Other COIN projects include  CosmoPhotoz (a Python galaxy photometric redshift estimator utilizing Generalized Linear Models) and CosmoABC (a Python package enabling Approximate Bayesian Computation for modeling astronomical data).  Please join us on GitHub!

+++++++++++++++++++++++++++++

Erzsebet Merenyi, organizer of new IEEE Task Force `Mining Complex Astronomical Data'

Astrominer TF : A Task Force has been founded for “Mining Complex Astronomical Data”, under the Data Mining Technical Committee of the IEEE Computational Intelligence Society.

The objective of the Astrominer TF is contribution to solutions of problems modern astronomy faces in turning the sky-full of inexhaustible stream of complex, high-dimensional data into reliable knowledge at an accelerated rate. Machine learning and data mining, computational intelligence (CI) approaches in general, are in high demand but as yet not sufficiently exploited in astronomy. We believe that meaningful contributions by experts in the relevant computational areas can only be achieved in tight collaboration with astronomers, and with the science needs driving the activities. It is equally important to make computational exerts  (engineers, computer scientists, statisticians, … ) aware of the existence of compelling scientific problems that require expert applications of state-of-the-art CI techniques and algorithms developed by the CI community, or motivate developments of new methods. The members of this TF strive to be directly involved in joint astronomical data analysis projects, publish joint papers, promote astronomy applications at IEEE and other CI type meetings and CI methodologies at astronomy meetings.

Several members contributed to the NOAO Workshop on Tools For Astronomical Big Data in Tucson, early March 2015.  More news are coming on-line in our web pages at http://astrominertf.joomla.com

+++++++++++++++++++++++++++++

Jessi Cisewski is lead organizer of the American Statistical Association Interest Group in Astrostatistics. 

The ASA astrostatistics interest group began in March 2014.  Information about the interest group – including how to join – can be found here.  

Astrostatistics is a very promising area of research for statisticians, and the interest group was formed to better facilitate and encourage communication and exchange of ideas between astrostatisticians along with drawing attention and credibility to the area among our peers. Statisticians are becoming more aware of the abundant, challenging, and intriguing problems that many of us have had the opportunity to explore.   We can instigate collaborations by providing a venue via the ASA to kindle interest in astrostatistics.  There are currently over 50 members in this interest group, and eventually we plan to turn the interest group into a full-fledged Section of the ASA.