# Questions tagged [canonical-correlation]

Canonical correlation analysis (CCA) is a method to analyze correlations between two sets of variables. It finds linear combinations of variables in each set such that their correlation is maximal.

132 questions
Filter by
Sorted by
Tagged with
26 views

### Getting sample/individual values from sparse CCA using PMA package

I have a dataset of 100 samples in which two sets of data were collected, 24950 features for one and 15 features in the other. I would like to apply sparse CCA to find correlations between these two ...
• 185
12 views

### Canonical Correlation Analysis CCA dimensionality reduction in R

I want to conduct dimensionality reduction using CCA. Let's suppose that $X$ is a matrix of $p$ predictors, $n$ samples and $y$ is a binary vector of $0$ and $1$, i.e. ...
• 249
7 views

### inverse of canonical correlation

The canonical correlation obtains the most correlative pairs of components across two matrices. However, the inverse is not commonly known but could be useful for obtaining the most different ...
• 73
16 views

### CCA in feature selection [R]

I'm trying to conduct feature selection based on the Canonical Correlation Analysis (CCA). As far as I know, for two given datasets $X$ and $Y$ CCA looks for the linear projections of $X$ and $Y$ such ...
• 249
19 views

### Self-Study: Score Statistic of a Poisson

I have been attempting a question on Generalised Linear Models. I thought I understood it but doing exercises myself now, proves otherwise. I have attached an image of the question here, struggling ...
6 views

### Canonical Correlation Analysis and Incorporating New Data

For some context, I am trying to follow this neuroscience paper (https://www.nature.com/articles/nm.4246?TB_iframe=true&width=921.6&height=921.6) Regarding my data, I have 2 datasets, one as a ...
• 185
1 vote
33 views

### Spread of species and site scores in R (vegan) DCA ordination

I have performed a Detrended Correspondence Analysis on social survey answers, aiming to see the spread of answers to 25 interview questions and how they related to the three farmer types (n =45) we ...
1 vote
23 views

### Multiview CCA with one source set and multiple independent target sets

I have 5 datasets from the same subjects that can be interpreted as different data modalities for these subjects. 4 of these datasets describe brain-related data and the last one contains behavioral ...
23 views

### Meaning of Canonical Correlation Analysis Orthogonal Constraint

Hello Folks - I'm trying to understand canonical correlation analysis (CCA). I found that it's useful for finding linear projections such that the projection results are maximally correlated. However, ...
19 views

### Canonical Correlation Analysis to combine two vector spaces

I have the following problem: I have two completely different vector spaces (namely word embedding vectors) from two different models. They do not have the same length since their vocabulary is ...
22 views

### How to train a model to maximize the difference of correlations?

I have two labeled datasets, $A$ and $B$: $(X_A, y_A)$ and $(X_B, y_B)$. $X_A, X_A \in \mathbb{R}^{m \times n}$ and $y_A, y_B \in \mathbb{R}^n$. $m$ is the number of features, $n$ is the number of ...
• 21
1 vote
97 views

For my current project I am using sklearn.cross_decomposition.CCA. On several wepages (e.g. https://stats.idre.ucla.edu/r/dae/canonical-correlation-analysis/ or ...
1 vote
15 views

### Why the correlation in CCA can be greater than 1? [closed]

I am working with the following paper https://home.ttic.edu/~klivescu/papers/andrew_icml2013.pdf where they use DCCA and in the results they present they show correlations between different sets of ...
• 11
1 vote
66 views

### Why is the first canonical direction equal to the left singular vector i.e. why is $w_1 = a = u_1$ in CCA (Canonical Correlation Analysis)?

I want to understand why the canonical direction $a$ is equal to the left singular value of $M = \Sigma^{-1/2}_X \Sigma_{X, Y} \Sigma^{-1/2}_Y$ and not $a = \sigma_1 u_1$. My calculation tell me that ...
• 5,886
27 views

### Maximization of correlation coefficient. The canonical correlation

The idea of the Canonical correlation is to maximize the correlation between two vectors $(Y,X)$ using 2 vector of weights $(\alpha,\beta)$. My problem is that I have a dependent variable $Y$ and a ...
40 views

5k views

### How to interpret results from Canonical Correlation Analysis (CCA)

I am learning CCA following an example posted here: https://stats.idre.ucla.edu/r/dae/canonical-correlation-analysis/ I have questions regarding on how to interpret the canonical coefficients, ...
• 131
I've tried reading many explanations of CCA, and I don't understand it. For example, on Wikipedia, it refers to two "vectors" $a$ and $b$ such that $\rho = \text{corr}(a^{\top} X, b^{\top} Y)$ is ...