Hypothesis ranking and the context of probabilities in an open-ended search

Details Hypothesis ranking and the context of probabilities in an open-ended search Hypothesis ranking and the context of probabilities in an open-ended search

Feb 1996

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=1996aph.....4..285b&link_type=abstract

ASTROPARTICLE PHYSICS, Volume 4, Issue 3, p. 285-291.

Physics

2

Scientific paper

In observational sciences such as astronomy, there are often no clear guidelines regarding potential source types or emission characteristics. As a consequence, data analyses often tend to be somewhat open-ended as they attempt to address a broad range of questions. In GeV and TeV gamma-ray astronomy, where analyses often deal with marginal significance levels, this may lead to biases in the assessment of an overall significance for a given finding unless an appropriate statistical penalty is paid to account for the total number of independant tests performed. Furthermore, many tests may not be amenable to immediate repetition with independant data either owing to experimental sensitivity, limited exposure or source duty cycle. The assignment of a statistical penalty is especially difficult in open-ended analyses where one must allow for arbitrary exploration of the data. Ideally, one would like to structure this penalty so as to give more weight to particular hypotheses that are a priori deemed more likely. This paper will propose methods to rank hypotheses and place chance probability calculations within the context of an exhaustive, open-ended search.

Affiliated with

Also associated with

No associations

Rating

Hypothesis ranking and the context of probabilities in an open-ended search 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 Hypothesis ranking and the context of probabilities in an open-ended search, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Hypothesis ranking and the context of probabilities in an open-ended search will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1153719