Non-independent components analysis

Geert Mesters and Piotr Zwiernik ()
Additional contact information
Piotr Zwiernik: https://www.upf.edu/web/econ/faculty/-/asset_publisher/6aWmmXf28uXT/persona/id/3421209

Economics Working Papers from Department of Economics and Business, Universitat Pompeu Fabra

Abstract: A seminal result in the ICA literature states that for AY=e, if the components of e are independent and at most one is Gaussian, then A is identified up to sign and permutation of its rows (Comon, 1994) In this paper we study to which extent the independence assumption can be relaxed by replacing it with restrictions on the moments or cumulants of e. We document minimal conditions for identifiability and propose efficient estimation methods based on the new identification results. In situations where independence cannot be assumed the efficiency gains can be significant relative to methods that rely on independence. The proof strategy employed highlights new geometric and combinatorial tools that can be adopted to study identifiability via higher order restrictions in linear systems.

Keywords: Independent components analysis; cumulants; moments; tensors; identification (search for similar items in EconPapers)
JEL-codes: C20 C30 (search for similar items in EconPapers)
Date: 2022-08
New Economics Papers: this item is included in nep-ecm
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
https://econ-papers.upf.edu/papers/1845.pdf Whole Paper (application/pdf)

Related works:
Working Paper: Non-Independent Components Analysis (2022)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:upf:upfgen:1845

Access Statistics for this paper

More papers in Economics Working Papers from Department of Economics and Business, Universitat Pompeu Fabra
Bibliographic data for series maintained by ( this e-mail address is bad, please contact ).