On the Accuracy of Error Propagation Calculations by Analytic Formulas Obtained for the Inverse Transformation
The accuracy of error propagation calculations is estimated for the transformation x → y = f(x) of the normally distributed random variable x. The estimation is based on the formulas for the error propagation obtained for the inverse transformation y → x of the normally distributed random variable y. In the general case, the calculation accuracy for the mean value and the variance of the random variable y is shown to be of the first order of magnitude in the variance of the random variable x.
D.J. Hudson. Lectures on Elementary Statistics and Probability (CERN, 1963).
G.G. Rode. Propagation of measurement errors and measured means of a physical quantity for the elementary functions cos x and arccos x. Ukr. J. Phys. 61, 345 (2016). https://doi.org/10.15407/ujpe61.04.0345
G.G. Rode. Propagation of the measurement errors and measured means of physical quantities for the elementary functions x^2 and vx. Ukr. J. Phys. 62, 184 (2017). https://doi.org/10.15407/ujpe62.02.0184
P. Kosobutsky. Analytical relations for the mathematical expectation and variance of a standard distributed random variable subjected to the vx transformation. Ukr. J. Phys. 63, 215 (2018). https://doi.org/10.15407/ujpe63.3.215