Multiclass Sparse Bayesian Regression for fMRI-Based Prediction

Table 2

Intrasubject analysis. Explained variance 𝜁 for the three different methods. The 𝑃-values are computed using a paired 𝑡-test. VB-MCBR yields the best prediction accuracy, while being more stable than the reference methods.

Methods	Mean 𝜁	Std 𝜁	P-val/Gibbs-MCBR
SVR	0.82	0.07	.0006***
Elastic net	0.9	0.02	.001***
BRR	0.92	0.02	.0358**
ARD	0.89	0.03	.0015***
Gibbs-MCBR	0.93	0.01	—
VB-MCBR	0.94	0.01	.99

**Level of significance of the 𝑃-values between 0.01 and 0.05.
***Level of significance of the 𝑃-values below 0.01.