Table 3.
Model comparison for linear mixed-effect logistic regression on alternative quantitative genetic models for survival of cross types.
| Model | df | AIC | BIC | Λ | χ2 |  |
 |
|---|---|---|---|---|---|---|---|
| μ0 + bS | 2 | 362.89 | 370.89 | -179.44 | |||
| μ0 + bS + bH | 3 | 364.69 | 376.70 | -179.34 | 0.1978 | 1 | 0.6565 |
| μ0 + bS + bH + bSS | 4 | 349.63 | 365.64 | -170.81 | 17.0631 | 1 | 0.0000 |
| μ0 + bS + bH + bSS + bHH* | 5 | 345.30 | 365.32 | -167.65 | 6.3258 | 1 | 0.0119 |
| μ0 + bS + bH + bSS + bHH + bSH | 6 | 345.49 | 369.51 | -166.74 | 1.8142 | 1 | 0.1780 |
Models are evaluated using Akaike's Information Criterion (AIC), Bayesian Information Criterion (BIC), and Log Likelihood (Λ) goodness-of-fit estimates. The best model is denoted by an asterisk. Chi-square and p-values compare the respective model to the simpler model in the row above.