.png){kind=icon}
Return
NONNDEPENDENT COMPONENTS ANALYSIS
.png){kind=icon}
PRE
AI
DOI：10.1214/24-AOS2373.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
A seminal result in the ICA literature states that for AY = s, if the components of s are independent and at most one is Gaussian, then A is identified up to sign and permutation of its rows (Signal Process. 36 (1994)). In this paper we study to which extent the independence assumption can be relaxed by replacing it with restrictions on higher order moment or cumulant tensors of s. We document new conditions that establish identification for several nonindependent component models, for example, common variance models, and propose efficient estimation methods based on the identification results. We show that in situations where independence cannot be assumed the efficiency gains can be significant relative to methods that rely on independence.
Keywords:
Independent component analysis
identifiability
cumulants
tensors
Journal
IF:
3.7
Papers:
2.8K
Citations:
2.9W
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveShareSaveA first application of independent component analysis to extracting structure from stock returns
OAAI
OAAIShareSaveShareSave
View more