1

I am trying to build sparse coefficient time series models but couldn't find a good resource/book to learn from. Can anyone share with me a good resource/book? It will be great if it comes with R codes.Can anyone use this kind model? Thanks.

Stephan Kolassa
  • 123,354
XingHua.Ma
  • 31
  • 1
    Could you be a little more explicit as to what you mean by "sparse" in this context and how this differs from the Box-Jenkins approach? – Stephan Kolassa Nov 21 '12 at 10:06
  • For example:if we build an ARMA(p,q) model,generally,we need estimate p+q coefficients.The sparse means some of the p+q coefficients may equal to zero.So the model's coefficients are sparse.Maybe I don't describe clearly or in time-series these models don't call sparse coefficient models.If like this what they named? Thanks. – XingHua.Ma Nov 21 '12 at 12:49

1 Answers1

1

Are you familiar with the Lasso, Elastic Net and similar regularization techniques? The Lasso specifically is used in regression with many regressors to yield a result where "many" cofficients are zero, which looks like what you are aiming for.

It looks like there is some interest out there in applying the Lasso and similar concepts to ARMA models. See, e.g., here or or here.

I would recommend that you look into regularization and google for terms like "regularization technique time series technique".

Stephan Kolassa
  • 123,354
  • I'd love to see an answer to the question posted at your first link... (http://stats.stackexchange.com/questions/10425/fitting-an-arimax-model-with-regularization-or-penalization-e-g-with-the-lasso) – Zach Nov 21 '12 at 16:14
  • 1
    @Zach: you and me both, it would make a lot of sense... But the paper behind my third link notes that simulations were done in R using lars and glmnet, so it might be worthwhile to mail the authors and ask for their code. – Stephan Kolassa Nov 21 '12 at 19:37
  • I failed to find the archived link of this - this is not working any more @StephanKolassa. – User1865345 Mar 09 '23 at 10:49
  • 1
    @User1865345: thanks for pointing this out. I don't know which paper I referred to back then (thank goodness I have learned a bit about better linking in the last 10 years). Since I get a malware warning, I will remove the link from the answer. – Stephan Kolassa Mar 09 '23 at 10:55
  • Thanks @StephanKolassa for updating. – User1865345 Mar 09 '23 at 10:56