Some Notes on the Precision of the Gradient of an Estimated Straight Line

In this article Dr Beech assembles some material concerning the confidence limits to be attached to the estimated gradients of linear functional relationships and compares them with those for the regression lines. He derives explicit formulae for these confidence limits, and also discusses the accuracy to which the relative error of x and y needs to be known when the gradient is being estimated. The more important formulae are summarised at the end and mathematical details are given in an appendix.