Inference---A Python Package for Astrostatistics

Details Inference---A Python Package for Astrostatistics Inference---A Python Package for Astrostatistics

Aug 2004

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2004head....8.4401l&link_type=abstract

American Astronomical Society, HEAD meeting #8, #44.01; Bulletin of the American Astronomical Society, Vol. 36, p.1208

Astronomy and Astrophysics

Astronomy

Scientific paper

Python is an object-oriented ``very high level language'' that is easy to learn, actively supported, and freely available for a large variety of computing platforms. It possesses sophisticated scientific computing capabilities thanks to ongoing work by a community of scientists and engineers who maintain a suite of open source scientific packages. Key contributions come from the STScI group maintaining PyRAF, a Python environment for running IRAF tasks. Python's main scientific computing packages are the Numeric and numarray packages implementing efficient array and image processing, and the SciPy package implementing a wide variety of general-use algorithms including optimization, root finding, special functions, numerical integration, and basic statistical tasks. We describe the Inference package, a collection of tools for carrying out advanced astrostatistical analyses that is about to be released as a supplement to SciPy. The Inference package has two main parts. First is a Parametric Inference Engine that offers a unified environment for analysis of parametric models with a variety of methods, including minimum χ2, maximum likelihood, and Bayesian methods. Several common analysis tasks are available with simple syntax (e.g., optimization, multidimensional exploration and integration, simulation); its parameter syntax is remensicent of that of SHERPA. Second, the package includes a growing library of diverse, specialized astrostatistical methods in a variety of domains including time series, spectrum and survey analysis, and basic image analysis. Where possible, a variety of methods are available for a given problem, enabling users to explore alternative methods in a unified environment, with the guidance of significant documentation. The Inference project is supported by NASA AISRP grant NAG5-12082.

Affiliated with

Also associated with

No associations

Rating

Inference---A Python Package for Astrostatistics does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Inference---A Python Package for Astrostatistics, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Inference---A Python Package for Astrostatistics will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1179610