The repr_pip computes PIP (Perceptually Important Points) representation from a time series.
Usage
repr_pip(x, times = 10, return = "points")
Arguments
x
the numeric vector (time series)
times
the number of important points to extract (default 10)
return
what to return? Can be important points ("points"),
places of important points in a vector ("places") or "both" (data.frame).
Value
the values based on the argument return (see above)
References
Fu TC, Chung FL, Luk R, and Ng CM (2008)
Representing financial time series based on data point importance.
Engineering Applications of Artificial Intelligence, 21(2):277-300