Astronomy and Astrophysics – Astrophysics
Scientific paper
Jun 1992
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=1992a%26a...259..386p&link_type=abstract
Astronomy and Astrophysics, Vol.259, NO. 1/JUNI, P. 386, 1992
Astronomy and Astrophysics
Astrophysics
Data Analysis, Numerical Methods
Scientific paper
The non parametric robust fitting method originally due to Theil is reviewed starting from the work of Sen (1968). In that paper only the estimate of the angular coefficient in a straight line fit is considered. Here it is shown that this method can be successfully employed to estimate also the intercept of a straight line, as well as to fit other functions such as polynomials and some non-linear curves: the Gaussian, the Lorentzian, the exponential and the power law. The general method scheme is retained, while a new formulation for the error estimator is proposed. The method proved to be unbiased and robust by the Monte-Carlo simulations, adopting the Gaussian and Cauchy error sources. Comparison with the standard methods are performed too. The suggested reliability of the Theil method to fit data with errors both in the dependent and independent variable is tested. The obtained results are not fully satisfactory.
No associations
LandOfFree
Revision of the Theil Fitting Method 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 Revision of the Theil Fitting Method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Revision of the Theil Fitting Method will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1199109