The creation and evaluation of a model to simulate the probability of conception in seasonal-calving pasture-based dairy heifers

Table 3 Results of discrimination tests for the models of heifer conception

Model	LR	GAM	MLR
Standard 50% threshold (%)
Sensitivity	97.20	100.00	99.14
Specificity	6.77	0.00	8.27
PPV	78.43	77.72	79.04
NPV	40.91	0.00	73.33
Accuracy	77.05	77.72	78.89
Matthews correlation coefficient	0.09	0.00	0.20
Optimal thresholds (%)	35.12	60.47	52.63
Sensitivity	99.78	99.57	98.71
Specificity	0.00	5.26	10.53
PPV	77.68	78.57	79.38
NPV	0.00	77.78	70.00
Accuracy	77.55	78.56	79.06
Matthews correlation coefficient	−0.02	0.17	0.21
AUC	0.66	0.64	0.71

Legend: LR logistic regression model, GAM generalised additive model, MLR mixed logistic regression model, PPV positive predictive value, NPV negative predictive value, AUC area under the ROC curve

ISSN: 2046-0481