Table 3.
Results of 2000 simulations.
| Sample size : n = 300 | ||||||
|---|---|---|---|---|---|---|
| (20% Cure, 40% RC) | (30% Cure, 40% RC) | (40% Cure, 60% RC) | ||||
| Bias | MSE | Bias | MSE | Bias | MSE | |
| EGG‐AFT mixture cure model | ||||||
| q | −0,048 | 0,096 | −0,013 | 0,077 | 0,050 | 0,210 |
| β 0 | −0,002 | 0,000 | −0,001 | 0,001 | 0,002 | 0,001 |
| β 1 | −0,000 | 0,001 | −0,000 | 0,001 | 0,001 | 0,003 |
| β 2 | 0,003 | 0,006 | 0,007 | 0,009 | 0,004 | 0,016 |
| β 3 | −0,003 | 0,003 | −0,003 | 0,004 | −0,005 | 0,008 |
| α 0 | −0,025 | 0,012 | −0,029 | 0,016 | −0,058 | 0,054 |
| α 1 | 0,011 | 0,019 | 0,006 | 0,019 | 0,008 | 0,034 |
| γ 0 | 0,398 | 2,435 | 0,052 | 0,104 | 0,124 | 0,518 |
| γ 1 | −0,298 | 2,111 | −0,031 | 0,148 | −0,100 | 0,492 |
| γ 2 | 0,024 | 0,028 | 0,006 | 0,008 | 0,014 | 0,012 |
| γ 3 | −0,054 | 1,007 | −0,034 | 0,433 | −0,059 | 0,650 |
| Likelihood ratio test | ||||||
| True value of q | q = 0 | q = 0.5 | q = 1 | |||
| Cov. H0 ≡ q = 0 | 6,50% | 48,40% | 79,70% | |||
| Cov. H0 ≡ q = 1 | 98,30% | 55,00% | 7,60% | |||
| EGG‐AFT model without cure | ||||||
| q | −1,514 | 2,375 | −2,125 | 4,604 | −2,195 | 5,010 |
| β 0 | −0,066 | 0,005 | −0,063 | 0,005 | −0,044 | 0,005 |
| β 1 | 0,028 | 0,002 | −0,024 | 0,003 | 0,161 | 0,035 |
| β 2 | 0,001 | 0,002 | 0,004 | 0,003 | 0,007 | 0,006 |
| β 3 | 0,000 | 0,001 | −0,003 | 0,002 | −0,005 | 0,004 |
| α 0 | 0,142 | 0,035 | 0,549 | 0,321 | 0,822 | 0,702 |
| α 1 | 0,241 | 0,083 | 0,020 | 0,033 | 0,220 | 0,092 |
RC, right‐censored; MSE, mean squared error; EGG‐AFT, extended generalized gamma accelarated failure time.
Bias and MSE of the EGG‐AFT mixture cure model in the upper part of the Table; rejection percentage of the likelihood ratio test in the middle; bias and MSE of the EGG‐AFT model in the lower part.