I am trying to detect a positively increasing trend in a time series data, to which when I apply the rolling mean it almost turns into a step wave function. The sharp rise in slope of rolling mean is an indication of positive trend, thus the rolling mean becomes like a step wave function. Is there a way or package to automatically detect the sharp increase in slope? -P [[alternative HTML version deleted]]
Possibly Parallel Threads
- multivariate, hierarchical model
- Simulation help
- Add trend line to XYPlot using a subset of the original data
- With the right kind of eyes you can almost see the high-water markāthat place where the wave finally broke and rolled back.
- How to find points of intersection between harmonic function and a line