Table 3.
Ma & Yin method | Naive method | Simex method | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Estimate | |||||||||||
200 | 0.1 | Bias | -2.1 | 1.3 | -2.2 | 3.8 | -10.6 | -2.1 | -1.4 | -0.3 | -2.2 |
Emp. | 12.6 | 21.2 | 6.7 | 11.2 | 16.2 | 6.7 | 12.4 | 20.6 | 6.7 | ||
Est. | 11.4 | 22.1 | 6.3 | 9.8 | 16.7 | 6.2 | 10.7 | 19.1 | 6.3 | ||
95% cv | 94.4 | 97.2 | 94.8 | 94.2 | 95.4 | 95.0 | 93.4 | 96.2 | 94.8 | ||
MSE | 12.6 | 21.2 | 6.8 | 11.4 | 17.3 | 6.7 | 12.4 | 20.6 | 6.8 | ||
200 | 0.25 | Bias | -5.6 | 9.3 | -2.6 | 19.9 | -42.9 | -1.9 | 7.4 | -18.6 | -2.0 |
Emp. | 18.5 | 45.0 | 7.1 | 9.4 | 10.0 | 6.7 | 12.1 | 20.7 | 6.8 | ||
Est. | 21.8 | 57.8 | 6.9 | 7.9 | 10.5 | 6.2 | 9.7 | 14.6 | 6.3 | ||
95% cv | 97.2 | 97.4 | 96.0 | 88.6 | 74.2 | 95.4 | 92.7 | 88.6 | 95.1 | ||
MSE | 18.8 | 45.9 | 7.2 | 13.3 | 28.4 | 6.7 | 12.6 | 24.2 | 6.8 | ||
300 | 0.1 | Bias | -3.1 | 2.5 | -1.3 | 2.5 | -9.4 | -1.2 | -2.5 | 1.3 | -1.3 |
Emp. | 10.0 | 16.7 | 4.3 | 9.2 | 12.8 | 4.3 | 9.8 | 16.4 | 4.3 | ||
Est. | 7.9 | 14.4 | 4.1 | 7.2 | 11.0 | 4.1 | 7.6 | 12.5 | 4.1 | ||
95% cv | 93.2 | 94.8 | 95.6 | 93.0 | 92.4 | 95.8 | 92.8 | 93.8 | 95.6 | ||
MSE | 10.1 | 16.8 | 4.4 | 9.3 | 13.6 | 4.3 | 9.9 | 16.4 | 4.4 | ||
300 | 0.25 | Bias | -6.8 | 10.5 | -1.9 | 18.7 | -41.2 | -1.1 | 6.0 | -15.8 | -1.4 |
Emp. | 15.3 | 37.1 | 4.7 | 7.4 | 8.2 | 4.3 | 9.9 | 17.4 | 4.4 | ||
Est. | 14.4 | 37.5 | 4.6 | 5.6 | 6.9 | 4.1 | 6.9 | 9.7 | 4.2 | ||
95% cv | 96.0 | 96.8 | 95.6 | 85.0 | 63.2 | 95.8 | 89.7 | 83.9 | 95.6 | ||
MSE | 15.8 | 38.2 | 4.7 | 10.9 | 25.2 | 4.3 | 10.3 | 19.9 | 4.4 |
Emp., empirical variance; Est., estimated variance; 95% cv, coverage probabilities of 95% confidence intervals computed based on the asymptotic normal distribution; MSE, mean squared error. All numbers were multiplied by 100.