Ivan Svetunkov

This is the default value in adam(). The additive trend model is more robust than the multiplicative one.

Some of the packages have now received strikes from CRAN, because they don't pass some of the checks on ATLAS / MKL. So, it would help a lot to package...

Done for `es()` in 0d119ec433cf6134fcc9565de76703591f20bc62

sma() could not be done as a wrapper, but now uses the adam() C++ code. Something similar should be done with ces, gum and ssarima.

Another group of functions to amend: 1. sim.es(), 2. sim.ssarima(), 3. sim.sma(), 4. sim.gum(), 5. sim.ces(), 6. sim.oes()

Given the difficulty of rewriting ces(), gum() and ssarima(), this is currently put on hold. Similar stuff applies for simulation functions. If someone is interested in doing that under my...

Now it makes sense to use ssgeneralfun.cpp instead

nus() can no rely on adamGeneral.cpp

This related to the project https://github.com/config-i1/greybox/projects/2

This is especially important for SSARIMA, for which the transition matrix can become quite ridiculous on high frequency data.