@pcarbo the BLiP method described here leads to better precision on the credible sets. I am not sure if there are any violations w.r.t coverage etc. But I think adding BLiP as a post-processing step with SuSiE can be useful any thoughts/comments on this?
@pcarbo the BLiP method described here leads to better precision on the credible sets. I am not sure if there are any violations w.r.t coverage etc. But I think adding BLiP as a post-processing step with SuSiE can be useful any thoughts/comments on this?