Table 3.

Some relevant paired comparisons.

	Mean MSE is significantly different and the difference (in %) is
Pair	>20%	[10%, 20%]	[0, 10%]
CMB better than HET	0	23	4
HET better than CMB	0	0	0
MOB better than CMB	12	1	2
CMB better than MOB	7	1	2
MOB better than GRF	13	1	0
GRF better than MOB	6	1	1
GRF better than HET	17	2	0
HET better than GRF	7	1	0

MSE: mean squared error.

Note: Paired-sample $t$ -tests comparing the mean MSE over the 100 repetitions are conducted at the 5% significance level. The results indicate the number of times one method significantly outperforms the other, along with the corresponding percentage difference in mean MSE, across all 27 scenarios.