Ratios of parameter estimates are often used in econometric applications. However, the test of these ratios when estimated can cause difficulties since the ratio of asymptotically normally distributed random variables have a Cauchy distribution for which there are no finite moments.
This paper presents a method for the estimation of confidence intervals based on the Fieller approach that has been shown to be preferable to the usual Delta method. Using example applications in both Stata and R, we demonstrate that a few extra steps in the examination of the estimate of the ratio may provide a confidence interval with superior coverage.