Table 3.
Ordinalized data simulation: Percent of simulated LRT statistics under the null hypothesis exceeding critical value for pairs of nested models based on 2000 replicates of N=1000.
| % Type I error rates | |||||||||
|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||
| DGM | LRT df | Ordinalized on M | Ordinalized on P | ||||||
| Model for HA | Model for H0 | Condition of H0 | 10 | 5 | 1 | 10 | 5 | 1 | |
| Cholesky with G×M (3) | NL main effects with G×M (4) | Large β2 | 4 | 4.9 | 2.4 | 0.3 | 3.0 | 1.0 | 0.2 |
| Small β2 | 4 | 4.4 | 1.8 | 0.1 | 1.9 | 0.7 | 0.2 | ||
| NL main effects (4*) | Large β2 | 7 | 5.9 | 2.6 | 0.6 | 4.4 | 2.2 | .25 | |
| Small β2 | 7 | 5.7 | 2.5 | 0.5 | 15.4 | 8.8 | 2.1 | ||
| Choleskya (2) | high rAM | 6 | 7.2 | 3.4 | 0.7 | 22.9 | 13.4 | 3.2 | |
| low rAM | 6 | 8.2 | 3.9 | 0.7 | 35.3 | 21.4 | 6.6 | ||
|
| |||||||||
| NL main effects with G×M (4) | NL main effects (4*) | Large β2 | 3 | 9.3 | 4.8 | 0.8 | 8.8 | 3.6 | 0.9 |
| Small β2 | 3 | 8.0 | 4.1 | 1.0 | 31.9 | 20.5 | 6.9 | ||
|
| |||||||||
| NL main effects (4*) | Lin main effectsa (4†) | 1 | 10.8 | 4.6 | 1.4 | 17.6 | 10.3 | 2.3 | |
|
| |||||||||
| Cholesky (2) | Lin Main effects (4†) | 2 | 8.3 | 4.4 | 0.7 | 9.4 | 4.8 | 0.7 | |
Note: P refers to a phenotype of interest, M refers to a putative moderator. LRT refers to the likelihood ratio test.
a
DGM does not contain G×M effects.