2020
DOI: 10.1038/s41598-020-69575-x
|View full text |Cite
|
Sign up to set email alerts
|

Multi-group analysis using generalized additive kernel canonical correlation analysis

Abstract: Multivariate analysis has been widely used and one of the popular multivariate analysis methods is canonical correlation analysis (CCA). CCA finds the linear combination in each group that maximizes the Pearson correlation. CCA has been extended to a kernel CCA for nonlinear relationships and generalized CCA that can consider more than two groups. We propose an extension of CCA that allows multi-group and nonlinear relationships in an additive fashion for a better interpretation, which we termed as Generalized… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
2
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
3
2

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(3 citation statements)
references
References 31 publications
0
2
0
Order By: Relevance
“…blood assays, behavior, genetics and neuroimaging). Yet, in addition to variations of CCA, ICA analyses have also been proposed to map across multiple variable domains, though Partial Least Squares (PLS; [82,122,123]) is perhaps the most common approach that allows mapping the combination of multiple data modalities into a single space [124,125]. Furthermore, the current work is also limited to linear interactions between data domains.…”
Section: Discussionmentioning
confidence: 99%
“…blood assays, behavior, genetics and neuroimaging). Yet, in addition to variations of CCA, ICA analyses have also been proposed to map across multiple variable domains, though Partial Least Squares (PLS; [82,122,123]) is perhaps the most common approach that allows mapping the combination of multiple data modalities into a single space [124,125]. Furthermore, the current work is also limited to linear interactions between data domains.…”
Section: Discussionmentioning
confidence: 99%
“…The contribution of a neuron to the new manifold is generally measured by correlation. Specifically, the contribution of ith neuron in delay1 phase is defined as follows (Bae et al, 2020): …”
Section: Methodsmentioning
confidence: 99%
“…While CCA is a powerful technique, it has some limitations, such as the assumptions of multivariate normality and strict regularity of intercorrelation matrices, sensitivity to outliers, and the need for a relatively large sample size for stable parameter estimation [ 34 , 35 ]. Therefore, in this study, we utilized a different approach called canonical analysis of covariance (CAC), also known as quasi-canonical analysis [ 36 , 37 ].…”
Section: Methodsmentioning
confidence: 99%