Abstract
According to the statistical law of large numbers, the expected mean of identically distributed random variables of a sample tends toward the actual mean as the sample increases. Under this law, it is possible to test the Chandrasekhar’s relation (CR), <V> = (pi/4)^{-1}<Vsini>, using a large amount of Vsini and V data from different samples of similar stars. In this context, we conducted a statistical test to check the consistency of the CR in the Kepler field. In order to achieve this, we use three large samples of V obtained from Kepler rotation periods and a homogeneous control sample of Vsin i to overcome the scarcity of Vsin i data for stars in the Kepler field. We used the bootstrap-resampling method to estimate the mean rotations (<V> and <Vsini>) and their corresponding confidence intervals for the stars segregated by effective temperature. Then, we compared the estimated means to check the consistency of CR, and analyzed the influence of the uncertainties in radii measurements, and possible selection effects. We found that the CR with <sini> = pi/4 is consistent with the behavior of the <V> as a function of <Vsini> for stars from the Kepler field as there is a very good agreement between such a relation and the data.
Author
J. R. P. Silva1, B. B. Soares, and D. B. de Freitas
Journal
ApJ
Paper Publication Date
November 2014
Paper Type
Astrostatistics
Paper
Silvaetal2014ApJ.pdf — 1848 KB