Table 1. Maximum likelihood estimates of the parameters in the Hurdle Negative Binomial model.
For infection intensity (hurdle model) | |||||
Parameter | Estimate | Std. Error | z value | Pr(>|z|) | |
Intercept | -1.8322 | 0.2750 | -6.66 | 2.7e-11 | *** |
LogGto | 0.9887 | 0.0577 | 17.15 | < 2e-16 | *** |
MOI | 0.2358 | 0.0970 | 2.43 | 0.0151 | * |
interaction | -0.0631 | 0.0212 | -2.97 | 0.0029 | ** |
Dispersion | 0.66957 | 0.027334 | |||
For infection prevalence (logit link function) | |||||
LogGto | 0.2440 | 0.0192 | 12.69 | < 2e-16 | *** |
MOI | -0.1574 | 0.0447 | -3.53 | 4.2e-04 | *** |
interaction | 0.0234 | 0.0133 | 1.76 | 0.078 | ns |
LogGto, logged-transformed gametocyte density; MOI, multipicity of infection; Interaction for interaction between LogGto and MOI. The dispersion parameter that is accommodated in the HNB model reflects over-dispersion of parasites. Z value is a Student statistics. Significant codes
'***' <0.001
'**' <0.01
'*' <0.05; 'ns' ≥0.05.