However, as Triquart et al. have argued, one should consider presenting the uncertainty around the point estimate above:
"SUCRA estimates are not currently presented with their uncertainty intervals. Because SUCRA is derived from the mean rank, the 95% CrI for mean rank should be translated into a 95% CrI for SUCRA"
Would it be possible to add this functionality to the multinma package?
Hi, if I understood the posterior_ranks() function correctly, SUCRA is only calculated as a point estimate (mean SUCRA):
sucras <- unname((ntrt - rk_summary$mean)/(ntrt - 1))However, as Triquart et al. have argued, one should consider presenting the uncertainty around the point estimate above:
"SUCRA estimates are not currently presented with their uncertainty intervals. Because SUCRA is derived from the mean rank, the 95% CrI for mean rank should be translated into a 95% CrI for SUCRA"
Would it be possible to add this functionality to the multinma package?
Thanks!