Graph network from correlation matrix
Hi!
I am new to network analysis. I am trying to create a graph network based on a correlation matrix that i have between 14 brain regions (positively or negatively correlated from resting state fMRI seeds).
Can i get the network analysis and visualization done without any groups or other data? Is it possible to cluster the network based on strength of connections?
Thank you for the help!
Comments
The problem seems to have been solved by adding another row and column to the table, that wasn't added to the dependent variable list.
Regarding the following error:
''The following problem(s) occurred while running the analysis:
The variance-covariance matrix of the supplied data is not positive-definite. Please check if variables have many missings observations or are collinear''
My correlation matrix does not have missing observation, but is based on correlations between 14 regions. How can i fix the error?
Thank you!
I've attended the relevant team member to this.
E.J.
The input has to be raw data and not a correlation matrix. Inputting a correlation matrix with an added row will give poor results (because that dataset will be treated as raw data).
Best, Sacha
Dear Sacha,
Thank you for the reply! I am new to this. The only output i have is a correlation matrix between various regions. How is it possible to convert to raw data? I attached the table.
Thank you!