Fig. 4
From: Flexible hidden Markov models for behaviour-dependent habitat selection
State decoding for zebra example, where blue points are encamped and orange points are exploratory. a Locations in each state, estimated via global decoding (Viterbi algorithm). b Estimated probability of being in the exploratory state (black line) and Viterbi sequence (coloured points) over a few days chosen for visualisation purposes (note separate y-axes)