. Author manuscript; available in PMC: 2022 Jan 1.

Published in final edited form as: Neurocomputing (Amst). 2020 Oct 14;422:235–244. doi: 10.1016/j.neucom.2020.10.009

Table 2:

The illustration of the relation between the overfitting risk and necessary sample size to achieve a particular statistical power using the male VAT data.

ROI	h_j	Training MSE	Test MSE	OR
[0.148, 0.281]	92	0.458	1.715	1.257
[0, 1]	233	0.412	2.381	1.969