As in a joint analysis, if the individual-level genotype data of the discovery sample are unavailable, we can estimate the LD correlations from the reference sample and approximate a conditional analysis as (21)b~2∣b~1=B2−1D2β^2−B2−1CB1−1D1β^1(22)var(b~2∣b~1)=σC2B2−1−σC2B2−1CB1−1C′B2−1 where B1 and B2 are similar as in equation (11) and C≈X2′X1 with the jkth element. (23)Cjk=min(n^2j,n^1k)2p2j(1−p2j)p1k(1−p1k)∑imw2ij2∑imw1ik2∑imw2ijw1ik
