Skip to main content
Figure 7 | EPJ Data Science

Figure 7

From: The shocklet transform: a decomposition method for the identification of local, mechanism-driven dynamics in sociotechnical time series

Figure 7

STAR. The Shocklet Transform And Ranking (STAR) algorithm combines the discrete shocklet transform (DST) with a series of transformations that yield intermediate results, such as the cusp indicator function (panel (C) in the figure) and windows during which each univariate time series displays shock-like behavior (panel (D) in the figure). Each of these intermediate results is useful in its own right, as we show in Sect. 2. We display the final output of the STAR algorithm, a univariate indicator that condenses information about which of the time series exhibits the strongest shock-like behavior at each point in time

Back to article page