I am trying to find the technique behind parameter estimation in ets function. Can anyone please suggest some good reading material to know how final alpha, beta and gamma is estimated and what search optimization logic is applied.
Thank you for the link. I would also like to know the optimization process behind the parameter estimation. So the question is does it solve like a linear constraint equations and comes with parameters value or it searches heuristically for parameters value which will be the most optimized possible numbers that will give maximum likelihood/minimum error metric. Want to know the internal working of parameter estimation or solving technique/optimization technique to come up with some numbers for alpha, beta and gamma.
It is a nonlinear constrained optimization using the Nelder-Mead algorithm. The code is open source so you can see for yourself how it is done.
Thank you so much for the direction. Thanks a lot!!