Thanks Stefan, yes, it was the little subtleties I was afraid of. I did have a look at the R implementation, which is actually a C implementation and thereby not that easy to read (for me). However, I will try to implement a basic version of the method over the next weekend.
Fabian On Tuesday, April 8, 2014 4:21:11 PM UTC+2, Stefan Karpinski wrote: > > From perusing the Wikipedia page, this technique looks straightforward to > implement using the linear algebra tools available. It seems like having a > good SVD implementation is the main difficulty – which we do. Perhaps there > are subtleties which aren't captured there but I suspect an R or Python > implementation could be ported in less than an hour. > > On Apr 8, 2014, at 9:49 AM, Fabian Gans <fabia...@gmail.com <javascript:>> > wrote: > > Hi all, > > has anyone implemented an SSA algorithm in julia or wrapped a library for > it? I could not find anything in the available packages and wanted to make > sure I didn't miss it. If not I will stick to R for this 😫. See here > http://en.wikipedia.org/wiki/Singular_spectrum_analysis for a description > of the algorithm. > > Cheers > Fabian > >