Table 1.

Phenotypic prediction of IQ from each algorithm in the training and test samples

Algorithm	r with IQ	SE lower	SE upper	p value
LASSO train	0.380	0.349	0.410	<2.2e–16
Ridge train	0.476	0.447	0.503	<2.2e–16
Shen train	0.322	0.290	0.353	<2.2e–16
LASSO test	0.188	0.172	0.205	<2.2e–16
Ridge test	0.212	0.195	0.228	<2.2e–16
Shen test	0.187	0.171	0.204	<2.2e–16

Phenotypic prediction of IQ by each algorithm, measured as the Pearson's r between brain-based predicted IQ and measured IQ. Training was done in the 3SK sample and testing in the 13SK sample. The top three columns represent the algorithm predicting in the sample it was trained in (with LASSO and ridge using 10-fold cross-validation to guard against overfitting); the bottom three represent the out-of-sample prediction going from the 3SK to the 13SK sample.