My question is which is the right way of computing those CI, if it is possible? I may try to divide the CI for the transition intensity in the numerator by the CI for the transition intensity in the denominator with the opposite order (so Lower/Upper and Upper/Lower), but is this correct? If not, is there another way?
Dear Prof. Jackson,
I have found at least a couple of papers where they show ratios of transition intensities with Confidence Intervals. See Table 3 of https://www.neurology.org/doi/10.1212/WNL.0000000000200051 or Table 2 of https://doi.org/10.1186/s12877-019-1291-z.
My question is which is the right way of computing those CI, if it is possible? I may try to divide the CI for the transition intensity in the numerator by the CI for the transition intensity in the denominator with the opposite order (so Lower/Upper and Upper/Lower), but is this correct? If not, is there another way?
Thank you in advance for your time!