Table 1 Simulation results for the high accuracy scenario
From: A double SIMEX approach for bivariate random-effects meta-analysis of diagnostic accuracy studies
Random-effects | ρ | \(\overline {\eta }\) | \(\overline {\xi }\) | \(\sigma ^{2}_{\eta }\) | \(\sigma ^{2}_{\xi }\) | ρ | Failure |
|---|---|---|---|---|---|---|---|
distribution | bias (s.e.) | bias (s.e.) | bias (s.e.) | bias (s.e.) | bias (s.e.) | rate % | |
Normal-Normal | |||||||
Normal | 0.2 | -0.19 (0.33) | 0.06 (0.23) | -0.47 (0.46) | -0.13 (0.23) | -0.04 (0.27) | 2.5 |
0.6 | -0.16 (0.33) | 0.03 (0.23) | -0.42 (0.47) | -0.12 (0.23) | -0.15 (0.22) | 1.8 | |
0.8 | -0.17 (0.34) | 0.02 (0.23) | -0.38 (0.48) | -0.10 (0.24) | -0.19 (0.18) | 1.8 | |
Binomial-Normal | |||||||
0.2 | 0.02 (0.37) | -0.01 (0.24) | -0.10 (0.67) | -0.03 (0.28) | 0.02 (0.36) | 6.2 | |
0.6 | 0.02 (0.38) | -0.01 (0.24) | -0.03 (0.71) | -0.02 (0.28) | -0.01 (0.28) | 11.1 | |
0.8 | 0.01 (0.37) | -0.00 (0.24) | -0.03 (0.69) | -0.01 (0.29) | -0.02 (0.20) | 19.5 | |
SIMEX | |||||||
0.2 | 0.08 (0.35) | -0.07 (0.25) | 0.05 (0.59) | 0.14 (0.30) | -0.03 (0.28) | 0.0 | |
0.6 | 0.08 (0.35) | -0.07 (0.25) | 0.08 (0.59) | 0.13 (0.29) | -0.13 (0.23) | 0.0 | |
0.8 | 0.05 (0.35) | -0.06 (0.25) | 0.07 (0.59) | 0.14 (0.30) | -0.18 (0.19) | 0.0 | |
Normal-Normal | |||||||
t | 0.2 | -0.22 (0.39) | 0.09 (0.28) | 0.07 (0.72) | 0.16 (0.38) | -0.06 (0.26) | 1.6 |
0.6 | -0.23 (0.41) | 0.04 (0.28) | 0.23 (0.80) | 0.18 (0.38) | -0.13 (0.22) | 0.7 | |
0.8 | -0.20 (0.41) | 0.03 (0.29) | 0.21 (0.78) | 0.22 (0.41) | -0.17 (0.17) | 0.8 | |
Binomial-Normal | |||||||
0.2 | 0.02 (0.46) | 0.01 (0.30) | 0.74 (1.13) | 0.34 (0.47) | -0.01 (0.33) | 3.6 | |
0.6 | -0.02 (0.48) | -0.01 (0.30) | 0.89 (1.20) | 0.36 (0.48) | -0.01 (0.24) | 7.9 | |
0.8 | 0.00 (0.47) | 0.00 (0.31) | 0.82 (1.20) | 0.40 (0.53) | -0.03 (0.18) | 14.0 | |
SIMEX | |||||||
0.2 | 0.05 (0.42) | -0.05 (0.31) | 0.64 (0.85) | 0.52 (0.47) | -0.05 (0.28) | 0.0 | |
0.6 | -0.01 (0.43) | -0.07 (0.31) | 0.77 (0.90) | 0.51 (0.47) | -0.10 (0.22) | 0.0 | |
0.8 | 0.00 (0.42) | -0.05 (0.31) | 0.66 (0.86) | 0.52 (0.47) | -0.15 (0.17) | 0.0 | |
Normal-Normal | |||||||
Skew-normal | 0.2 | -0.64 (0.29) | 0.17 (0.22) | -0.54 (0.38) | -0.13 (0.22) | 0.03 (0.26) | 1.4 |
(low skewness) | 0.6 | -0.57 (0.30) | 0.00 (0.23) | -0.49 (0.40) | -0.13 (0.23) | -0.11 (0.21) | 1.8 |
0.8 | -0.52 (0.31) | -0.10 (0.23) | -0.44 (0.43) | -0.13 (0.22) | -0.17 (0.17) | 2.1 | |
Binomial-Normal | |||||||
0.2 | -0.54 (0.32) | 0.12 (0.23) | -0.35 (0.48) | -0.05 (0.26) | 0.10 (0.33) | 5.4 | |
0.6 | -0.49 (0.32) | -0.05 (0.24) | -0.28 (0.53) | -0.03 (0.28) | 0.03 (0.26) | 10.6 | |
0.8 | -0.44 (0.34) | -0.16 (0.24) | -0.18 (0.59) | -0.02 (0.29) | 0.00 (0.19) | 20.0 | |
SIMEX | |||||||
0.2 | -0.48 (0.32) | 0.06 (0.24) | -0.16 (0.48) | 0.11 (0.28) | 0.05 (0.27) | 0.0 | |
0.6 | -0.42 (0.33) | -0.11 (0.25) | -0.10 (0.51) | 0.13 (0.29) | -0.09 (0.22) | 0.0 | |
0.8 | -0.38 (0.33) | -0.22 (0.25) | -0.04 (0.54) | 0.13 (0.30) | -0.16 (0.18) | 0.0 | |
Normal-Normal | |||||||
Skew-normal | 0.2 | -0.59 (0.30) | 0.37 (0.20) | -0.52 (0.40) | -0.19 (0.18) | 0.20 (0.23) | 1.2 |
(high skewness) | 0.6 | -0.44 (0.31) | 0.24 (0.22) | -0.44 (0.43) | -0.15 (0.21) | 0.01 (0.18) | 1.8 |
0.8 | -0.32 (0.32) | 0.15 (0.22) | -0.43 (0.44) | -0.13 (0.21) | -0.11 (0.15) | 1.2 | |
Binomial-Normal | |||||||
0.2 | -0.49 (0.32) | 0.35 (0.21) | -0.30 (0.51) | -0.13 (0.21) | 0.30 (0.29) | 7.6 | |
0.6 | -0.33 (0.35) | 0.22 (0.23) | -0.14 (0.62) | -0.06 (0.26) | 0.16 (0.20) | 15.1 | |
0.8 | -0.19 (0.37) | 0.13 (0.24) | -0.08 (0.66) | -0.04 (0.27) | 0.05 (0.16) | 25.6 | |
SIMEX | |||||||
0.2 | -0.43 (0.32) | 0.31 (0.21) | -0.13 (0.50) | -0.02 (0.22) | 0.22 (0.24) | 0.0 | |
0.6 | -0.26 (0.34) | 0.17 (0.23) | 0.00 (0.55) | 0.06 (0.26) | 0.03 (0.19) | 0.0 | |
0.8 | -0.14 (0.35) | 0.08 (0.24) | 0.02 (0.57) | 0.09 (0.27) | -0.10 (0.16) | 0.0 | |