Statistics – Computation
Scientific paper
Feb 1988
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=1988a%26as...72..171b&link_type=abstract
Astronomy and Astrophysics Supplement Series (ISSN 0365-0138), vol. 72, no. 2, Feb. 1988, p. 171-192. Research supported by the
Statistics
Computation
3
Astrolabes, Binary Stars, Error Analysis, Radio Stars, Stellar Evolution, Astronomical Models, Computational Astrophysics, Least Squares Method
Scientific paper
An analysis of the Paris astrolabe reductions is carried out. By use of "influential diagnostic" statistics, the author identifies star observations that appear to have a disproportionate influence on the estimated model of reduction and thus appear to be suspect or atypical. He discusses their strong influence - in the reductions - on the estimated parameters, error variance, internal corrections and nonnormality of the residual distributions. Finally, some advice is given to overcome some difficulties entailed by the suspicious data in the current reductions of the observational programme, since suspicious data may also reveal some lack of fit of the usual reduction model.
No associations
LandOfFree
Nonnormality of the errors and suspicious observations - Application to the Paris-astrolabe programme 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 Nonnormality of the errors and suspicious observations - Application to the Paris-astrolabe programme, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Nonnormality of the errors and suspicious observations - Application to the Paris-astrolabe programme will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1224533