Fig. 4
From: Proximate cues to phases of movement in a highly dispersive waterfowl, Anas superciliosa
Partial dependence plots for three important variables for predicting initiation of an exploratory phase. a Partial dependence on daily maximum temperature. b Partial dependence on minimum daily atmospheric pressure. c Partial dependence on total weekly rainfall at a bird’s location. Partial dependence is the dependence of the probability of change on one predictor variable after averaging out the effects of the other predictor variables in the model. The Y-axis is half the logit-probability of an individual transitioning to a phase of exploratory behaviour (see Cutler et al. [64] for detailed explanation)