It is my pleasure to tell you all that I have recently released UComp (at last!), a R package or MATLAB/Octave toolbox to run Unobserved Components models AUTOMATICALLY. That means that you do not have to assume any model because UComp suggests one to you based on information criteria (cool!).

You can install it from CRAN in the usual way (in R, here is the webpage if you want to see the manual) or download it from GitHub (for MATLAB/Octave, here also with a manual). By the way, you might also would like to have a look at my other projects, like SSpace or ECOTOOL.

There is also an example for UComp, sort of a vignette here for R users. Though, as a starter you may try the code at the end of this post.

Since the core functions are written in C++, just by doing some wrappers it can be «linked» to other usual «languages» (well, ok, I admit it is not so easy in the end). As a matter of fact, any contributor willing to do such wrappers in Python or any other language (Go, Julia, …) is very, very welcome, just let me know and we may have a word.

Just to start to use it try this:

install.packages(“UComp”)

library(UComp)

m = UC(log(AirPassengers))

plot(m$comp)

The same code in MATLAB/Octave would be:

load air passengers

M = UC(log(y), frequency);

plot(m.comp)

Enjoy!!