Table 4. Prediction performance for the PstvC dataset.

	RMSE
Node i	MA(1)	ES(0.89)	NN+BU	NN+MinT	NN+SR(0.4, 2.4)
Root	2.69	2.69	2.90 ± 0.05	2.58 ± 0.07	2.49 ± 0.03
2	1.20	1.20	1.33 ± 0.03	1.16 ± 0.03	1.12 ± 0.01
3	1.49	1.42	1.12 ± 0.01	1.16 ± 0.03	1.27 ± 0.02
4	1.11	1.11	1.25 ± 0.03	1.63 ± 0.02	1.06 ± 0.01
Mid-level	1.27	1.24	1.23 ± 0.01	1.32 ± 0.03	1.15 ± 0.01
5	0.53	0.53	0.56 ± 0.01	0.50 ± 0.01	0.53 ± 0.01
6	0.49	0.49	0.55 ± 0.02	0.48 ± 0.01	0.51 ± 0.02
7	0.46	0.46	0.49 ± 0.01	0.46 ± 0.01	0.43 ± 0.01
8	0.56	0.55	0.48 ± 0.01	0.48 ± 0.01	0.54 ± 0.03
9	0.57	0.54	0.42 ± 0.01	0.46 ± 0.02	0.53 ± 0.04
10	0.49	0.47	0.43 ± 0.01	0.42 ± 0.01	0.45 ± 0.01
11	0.48	0.48	0.55 ± 0.02	0.51 ± 0.01	0.49 ± 0.02
12	0.59	0.57	0.51 ± 0.01	0.48 ± 0.01	0.55 ± 0.02
13	0.44	0.44	0.45 ± 0.02	0.42 ± 0.01	0.43 ± 0.01
Bottom-level	0.51	0.50	0.49 ± 0.00	0.47 ± 0.01	0.50 ± 0.00
Average	0.85	0.84	0.85 ± 0.00	0.83 ± 0.01	0.80 ± 0.00