Error (or uncertainty) propagation is the practice of analyzing and accounting for the effect of numeric quantities' uncertainties on the results of functions that involve them.

When variables used in a function or mathematical operation have errors (due to measurement uncertainties, random fluctuations, sample variance, etc.), error propagation can be used to determine the resulting error of the function's output.