Home>

There are two matrices 300 by 300 elements. On the Y-axis the name of the program, on the X-axis of their version. How to properly correlate between them and what python tools can be used. I've tried multiple linear regression, but it doesn't really fit here. What other options are there besides regression.

Examples of matrices.

Matrix 1:

Version1 Version2 Version3
Program1 0.7865 0.6438 0.5323
Program2 0.0235 0.6327 0.7324
Program3 0.0843 0.5362 0.7533

Matrix 2:

Version1 Version2 Version3
    Program1 0.5656 0.8534 0.9246
    Program2 0.0325 0.8424 0.8265
    Program3 0.6391 0.2567 0.1577

For example, I took the size 3 by 3, the point is that I need to find the correlation between the data of these two matrices

With whom do you want to correlate "them"? Between software versions? What about regression? What is the independent variable in the regression and what is the dependent?

passant2021-10-13 07:24:33

Yes, between software versions. The regression did not fit here precisely for this reason, because it is impossible to determine the dependent and independent variable

Anna2021-10-13 07:31:51

Show at least an example of your data, so nothing is clear. And what the original problem sounds like, otherwise it seems to me that you are solving the wrong problem that you have.

CrazyElf2021-10-13 08:14:17

Added examples

Anna2021-10-13 08:48:10

Still, the correlation is sought between two series of numbers. Do you want to find the correlation between the same versions in two matrices? Well, then iterate over the columns, take the same column from matrix1 and matrix2 and read the correlation coefficient. Or it is the same with rows, if correlation is needed between one program in different matrices. And there are a million different methods for calculating correlations: realpython.com/numpy-scipy-pandas-correlation-python

CrazyElf2021-10-13 09:25:34