In this case, the off-diagonal elements of ∑j can be set equal to zero, so LD score regression needs to be run only T rather than T(T+1)/2 times. Note that this version of MTAG does not take into account correlation in estimation error across traits that is due to bias. For this reason, LD score regression should be run on the MTAG results, and the resulting MTAG standard errors should be inflated by the square root of the estimated intercept.
