Personal communications with Rebecca Killick r.killick@lancaster.ac.uk
"If you take the first differences of the data then you now have a 0-mean process and the problem reduces to a change in variance, i.e. cpt.var(gcc,method='PELT',penalty='SIC1'). I noticed in your code that you use smoothing splines on the data first. This is not advised as it changes the structure of the data and makes abrupt changes appear more gradual."
Personal communications with Rebecca Killick r.killick@lancaster.ac.uk
"If you take the first differences of the data then you now have a 0-mean process and the problem reduces to a change in variance, i.e. cpt.var(gcc,method='PELT',penalty='SIC1'). I noticed in your code that you use smoothing splines on the data first. This is not advised as it changes the structure of the data and makes abrupt changes appear more gradual."