minor issue on documenation on where changepoints

[stephens999]

thanks for the very nice package!

I was a little mislead by the statement "A changepoint is denoted as the first observation of the new segment / regime." in ?cpt.mean

I interpreted this as meaning that a changepoint at 100 means that there is a change between 99 and 100. But it seems that in fact it means there is a change between 100 and 101?

[rkillick]

This has been corrected to "A changepoint is the last observation of the current segment" (not sure when as I think that I did this and forgot to close this issue - silly me!).
This means that if you do c(rnorm(100),rnorm(100,5)) as the data then the changepoint you expect to get is 100 and the change is between observations 100 and 101.