Table 3 Performances of multiple testing at \(\alpha =0.05\), where \(n_1,\ldots ,n_G\) are set as \(n_1=\cdots =n_G=m\).
From: Debiased inference for heterogeneous subpopulations in a high-dimensional logistic regression model
Testing | Covariate | m | p | Power | FWER | ||||
|---|---|---|---|---|---|---|---|---|---|
DFGL | DL | DL-E | DFGL | DL | DL-E | ||||
Homogeneity | AR(1) | 200 | 80 | 0.408 (0.253) | 0.170 (0.197) | 0.015 (0.069) | 0.000 | 0.190 | 0.000 |
300 | 120 | 0.618 (0.229) | 0.358 (0.208) | 0.120 (0.172) | 0.000 | 0.060 | 0.020 | ||
Block | 200 | 80 | 0.395 (0.291) | 0.178 (0.231) | 0.045 (0.097) | 0.000 | 0.140 | 0.000 | |
300 | 120 | 0.690 (0.236) | 0.380 (0.245) | 0.085 (0.139) | 0.010 | 0.140 | 0.000 | ||
Significance | AR(1) | 200 | 80 | 0.649 (0.131) | 0.518 (0.104) | 0.352 (0.048) | 0.000 | 0.030 | 0.000 |
300 | 120 | 0.791 (0.092) | 0.684 (0.090) | 0.428 (0.088) | 0.000 | 0.010 | 0.000 | ||
Block | 200 | 80 | 0.690 (0.141) | 0.501 (0.110) | 0.377 (0.067) | 0.010 | 0.030 | 0.000 | |
300 | 120 | 0.841 (0.110) | 0.677 (0.116) | 0.421 (0.085) | 0.010 | 0.020 | 0.000 | ||
