References & Citations
Mathematics > Statistics Theory
Title: Joint estimation of intersecting context tree models
(Submitted on 3 Feb 2011 (v1), last revised 3 Oct 2012 (this version, v3))
Abstract: We study a problem of model selection for data produced by two different context tree sources. Motivated by linguistic questions, we consider the case where the probabilistic context trees corresponding to the two sources are finite and share many of their contexts. In order to understand the differences between the two sources, it is important to identify which contexts and which transition probabilities are specific to each source.
We consider a class of probabilistic context tree models with three types of contexts: those which appear in one, the other, or both sources. We use a BIC penalized maximum likelihood procedure that jointly estimates the two sources.
We propose a new algorithm which efficiently computes the estimated context trees. We prove that the procedure is strongly consistent. We also present a simulation study showing the practical advantage of our procedure over a procedure that works separately on each dataset.
Submission history
From: Aurélien Garivier [view email][v1] Thu, 3 Feb 2011 13:12:16 GMT (28kb)
[v2] Tue, 25 Oct 2011 15:09:58 GMT (26kb)
[v3] Wed, 3 Oct 2012 14:45:41 GMT (25kb)
Link back to: arXiv, form interface, contact.