Table 4. Post-hoc Tukey test for Public internal testing data. In the analysis of mean absolute error (MAE) for Public internal testing data, no significant difference was observed between the Combined model (CM) and the Public model (PM).
|
Post-hoc, Tukey test for Public test data | ||||||
|
Group 1 |
Group 2 |
MAE difference (months) |
P value |
Lower (months) |
Upper (months) |
Significant |
|
BM MAE |
CM MAE |
−5.50 |
<0.001 |
−6.97 |
−4.03 |
Yes |
|
BM MAE |
PM MAE |
−6.28 |
<0.001 |
−7.75 |
−4.81 |
Yes |
|
CM MAE |
PM MAE |
−0.78 |
0.43 |
−2.24 |
0.68 |
No |