The middle panel of Table 1 compares the power of competing methods under the situation that the SNP has the same level of association with each of the phenotypic traits: β=(0.3,0.3,0.3,0.3,0.3)′. The power of MANOVA decreased rapidly as the correlation ϱ increased. When ϱ=0.75, for instance, all the other methods had the power of at least 0.4 but the power of MANOVA was only 0.2. Further, PCA and TATES had higher power than MANOVA when ϱ>0. Yet, none of these three methods can beat the three Fisher combination tests (FC-Permutation, FC-Pearson, and FC-Kendall) that performed equally well.
