Open ernesto561 opened 1 year ago
I didn't have any source code for this implementation. So, I did it on my own. Actually, there were some unknowns for me in Rossi et al. (2017) related to the missing threshold parameter alpha in equation (5) and how exactly the moving windows were defined computationally.
In the documentation for the function
thresh
, it's mentioned that the use ofmethod = "NLS"
is not recommended. Is this related to the implementation of the method in the package or to the method itself? By the way, I have tried to use the NLS method with different datasets and always gett=0
. Thanks.