Table 4 Model performance of four regression methods that map FACT- L to SF- 6D utility scores
From: Prediction of the SF-6D utility score from Lung cancer FACT-L: a mapping study in China
No | Mapping | RMSE | MAE | CCC | AEโ>โ0.1 | AEโ>โ0.05 | AIC | BIC | ARV |
|---|---|---|---|---|---|---|---|---|---|
| ย | method | ย | ย | ย | (%) | (%) | ย | ย | ย |
1 | OLS M1 | 0.0977 | 0.0764 | 0.7482 | 58.24 | 29.92 | -1129.08 | -1120.20 | 4.64 |
2 | OLS M2 | 0.0977 | 0.0763 | 0.7485 | 57.92 | 29.44 | -1127.63 | -1114.31 | 4.36 |
3 | OLS M3 | 0.0849 | 0.0658 | 0.8211 | 55.20 | 21.28 | -1297.23 | -1270.60 | 2.50 |
4 | OLS M4 | 0.0841 | 0.0654 | 0.8251 | 53.12 | 21.60 | -1299.07 | -1250.25 | 2.00 |
5 | OLS M5 | 0.0838 | 0.0653 | 0.8265 | 53.60 | 21.28 | -1299.42 | -1241.73 | 1.50 |
6 | TOBIT M1 | 0.0980 | 0.0763 | 0.7558 | 57.76 | 30.08 | -888.24 | -874.92 | 4.14 |
7 | TOBIT M2 | 0.0981 | 0.0765 | 0.7555 | 57.92 | 29.92 | -886.38 | -868.63 | 4.86 |
8 | TOBIT M3 | 0.0851 | 0.0658 | 0.8250 | 54.08 | 19.84 | -1035.70 | -1004.64 | 2.43 |
9 | TOBIT M4 | 0.0844 | 0.0655 | 0.8286 | 53.60 | 21.44 | -1037.33 | -984.08 | 1.93 |
10 | TOBIT M5 | 0.0841 | 0.0655 | 0.8300 | 52.96 | 21.60 | -1038.11 | -975.99 | 1.64 |
11 | OPROBIT M1 | 0.0975 | 0.0761 | 0.7519 | 59.20 | 28.96 | 8089.15 | 14211.95 | 4.86 |
12 | OPROBIT M2 | 0.0973 | 0.0760 | 0.7529 | 57.60 | 29.44 | 8088.51 | 14151.67 | 4.14 |
13 | OPROBIT M3 | 0.0852 | 0.0662 | 0.8204 | 53.92 | 20.32 | 7570.70 | 13350.07 | 2.57 |
14 | OPROBIT M4 | 0.0843 | 0.0657 | 0.8245 | 54.40 | 21.92 | 7567.43 | 13290.18 | 2.00 |
15 | OPROBIT M5 | 0.0839 | 0.0655 | 0.8262 | 52.96 | 21.76 | 7568.51 | 13309.17 | 1.43 |
| ย | Beta-mixture regression models without truncation | ย | ย | ย | ย | ย | |||
16 | BETAMIX M1a | 0.0973 | 0.0768 | 0.7500 | 58.24 | 30.24 | -476.33 | -454.14 | 9.93 |
17 | BETAMIX M1b | 0.0980 | 0.0724 | 0.7312 | 59.36 | 29.76 | -500.06 | -460.12 | 10.00 |
18 | BETAMIX M2a | 0.0973 | 0.0721 | 0.7506 | 58.56 | 29.60 | -474.50 | -443.44 | 9.50 |
19 | BETAMIX M3a | 0.0847 | 0.0657 | 0.8225 | 54.40 | 21.12 | -628.08 | -570.39 | 3.64 |
20 | BETAMIX M3b | 0.0848 | 0.0661 | 0.8217 | 54.24 | 21.28 | -665.21 | -572.02 | 4.07 |
21 | BETAMIX M3c | 0.0850 | 0.0663 | 0.8219 | 54.88 | 21.92 | -672.91 | -544.21 | 5.00 |
22 | BETAMIX M4a | 0.0842 | 0.0654 | 0.8250 | 53.92 | 21.12 | -620.49 | -518.42 | 2.86 |
23 | BETAMIX M5a | 0.0948 | 0.0653 | 0.8265 | 54.08 | 21.76 | -624.50 | -504.69 | 3.86 |
| ย | Beta-mixture regression models with truncation | ย | ย | ย | ย | ย | |||
24 | BETAMIX M1a# | 0.1585 | 0.1213 | 0.3480 | 72.16 | 47.84 | 155.21 | 168.52 | 13.43 |
25 | BETAMIX M2a# | 0.1585 | 0.1211 | 0.3480 | 72.00 | 47.68 | 154.95 | 172.70 | 13.00 |
26 | BETAMIX M2b# | 0.0968 | 0.0763 | 0.7549 | 58.72 | 29.44 | -542.86 | -489.61 | 8.43 |
27 | BETAMIX M3a# | 0.0851 | 0.0660 | 0.8226 | 54.24 | 20.96 | -589.12 | -531.43 | 4.50 |
28 | BETAMIX M4a# | 0.1538 | 0.1206 | 0.3992 | 75.68 | 48.48 | 44.94 | 98.21 | 12.57 |
29 | BETAMIX M5a# | 0.0843 | 0.0656 | 0.8263 | 53.92 | 22.40 | -589.44 | -469.62 | 4.21 |