Table 1 MSE performance of different methods in simulated datasets after adjusting for mean imputation performance
From: Dynamic model updating (DMU) approach for statistical learning model building with missing data
Settings | p | Average (MSE (Method)/MSE (Mean Imputation) (S = 30) | ||||
|---|---|---|---|---|---|---|
SLR (95% CI) | KNN (95% CI) | SLRM (95% CI) | RF (95% CI) | DMU (95% CI) | ||
SCR | 20 | 1.03 (0.9–1.17) | 1 (0.88–1.12) | 2.03 (1.81–2.26) | 1.17 (1.06–1.28) | 0.99 (0.88–1.1) |
|  | 25 | 1.2 (1.05–1.35) | 0.97 (0.86–1.08) | 2.01 (1.8–2.22) | 1.12 (1.02–1.22) | 1.08 (0.97–1.2) |
|  | 30 | 1.44 (1.24–1.63) | 0.98 (0.86–1.09) | 1.95 (1.77–2.13) | 1.1 (1.01–1.18) | 0.98 (0.90–1.06) |
NCR | 20 | – | – | 1.89 (1.69–2.1) | 1.1 (1.01–1.2) | 1.29 (1.19–1.38) |
|  | 25 | – | – | 2.07 (1.85–2.29) | 1.12 (1.03–1.22) | 1.59 (1.47–1.71) |
|  | 30 | – | – | 1.83 (1.62–2.04) | 1.05 (0.96–1.13) | 1.73 (1.6–1.87) |