CI diagrams for different segments of the blazar light curve for different embedding dimensions, same as in Fig. 2: the last 3 digits of the dataset (see Table 1) are indicated. The upper panel of each box is for the real dataset, the lower one is for the phase-randomized surrogate. None of the diagrams shows clear indication for a strange attractor of low dimension, especially considering that the “surrogate” diagrams are very similar to the ones of the real data.